From Servitude to Service Work: Historical Continuities in the Racial Division of Paid Reproductive Labor Author(s): Evelyn Nakano

Glenn Source: Signs, Vol. 18, No. 1 (Autumn, 1992), pp. 1-43 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/3174725 . Accessed: 15/04/2011 17:41
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=ucpress. . Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Signs.

http://www.jstor.org

FromServitude to ServiceWork: in the Racial HistoricalContinuities Division of PaidReproductive Labor

Evelyn
R

Nakano

Glenn

E C E N T SCHO L A R S H I P on African American, Latina,

Asian American, and Native American women reveals the complex interactionof race and gender oppression in their lives. These studiesexpose the inadequacyof additivemodels that treat genderand race as separateand discretesystemsof hierarchy (Collins 1986; King 1988; Brown 1989). In an additive model, white women are viewed solely in terms of gender,while women of color are effectsof gender by the cumulative thoughtto be "doubly"subordinated Yet a more race. one that captures adequateframework, plus achieving the interlocking, interactive natureof these systems,has been extraordirace and genderhave developedas separate narilydifficult.Historically, of own literature and concepts.Thus features each with its topics inquiry, of social life consideredcentralin understanding one system have been overlookedin analysesof the other. One domainthat has been exploredextensivelyin analysesof gender but ignored in studies of race is social reproduction.The term social is used by feministscholarsto referto the arrayof activities reproduction and relationshipsinvolved in maintainingpeople both on a daily basis and intergenerationally. Reproductivelabor includes activities such as household and servingfood, laundering and purchasing goods, preparing and repairingclothing, maintainingfurnishings appliances,socializing children,providingcareand emotionalsupportfor adults,and maintaining kin and communityties.
Workon this projectwas madepossibleby a Title F leavefrom the StateUniversity of New Yorkat Binghamton and a visitingscholarappointment at the MurrayResearch Centerat RadcliffeCollege.Discussionswith ElsaBarkley Brown,GaryGlenn,Carole to the ideas developedhere.My Turbin,and BarrieThornecontributed immeasurably thanksto Joyce Chinenfor directing me to archivalmaterials in Hawaii.I am also grateful to members of the Womenand WorkGroupand to Norma Alarcon,GaryDymski, AntoniaGlenn,MargaretGuilette,Terence Hopkins,EileenMcDonagh,JoAnnePreston, for theirsuggestions. Mary Ryan,and four anonymousSignsreviewers
[Signs:Journal of Womenin Cultureand Society 1992, vol. 18, no. 1] ? 1992 by The Universityof Chicago. All rights reserved.0097-9740/93/1801-0007$01.00

Autumn

1992

SIGNS

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Marxist feminists place the gendered construction of reproductive labor at the center of women's oppression. They point out that this labor is performed disproportionately by women and is essential to the industrial economy. Yet because it takes place mostly outside the market, it is invisible, not recognized as real work. Men benefit directly and indirectly from this arrangement-directly in that they contribute less labor in the home while enjoying the services women provide as wives and mothers and indirectly in that, freed of domestic labor, they can concentrate their efforts in paid employment and attain primacy in that area. Thus the sexual division of reproductive labor in the home interacts with and reinforces sexual division in the labor market.1 These analyses draw attention to the dialectics of production and reproduction and male privilege in both realms. When they represent gender as the sole basis for assigning reproductive labor, however, they imply that all women have the same relationship to it and that it is therefore a universal female experience.2 In the meantime, theories of racial hierarchy do not include any analysis of reproductive labor. Perhapsbecause, consciously or unconsciously, they are male centered, they focus exclusively on the paid labor market and especially on male-dominated areas of production.3 In the 1970s several writers seeking to explain the historic subordination of peoples of color pointed to dualism in the labor market-its division into distinct markets for white workers and for racial-ethnic workers-as a major vehicle for maintaining white domination (Blauner 1972; Barrera 1979).4 According to these formulations, the labor system has been organized to ensure that racial-ethnic workers are relegated to a lower tier of low-wage, dead-end, marginal jobs; institutional barriers, including restrictions on legal and political rights, prevent their moving out of that tier and competing with Euro-American workers for better jobs. These theories draw attention to the material advantages whites gain from the racial division of labor. How1 For various formulations, see Benston (1969), Secombe (1974), Barrett (1980), Fox (1980), and Sokoloff (1980). 2 Recently, white feminists have begun to pay attention to scholarship by and about racial-ethnic women and to recognize racial stratification in the labor market and other public arenas. My point here is that they still assume that women's relationship to domestic labor is universal; thus they have not been concerned with explicating differences across race, ethnic, and class groups in women's relationship to that labor. 3 See, e.g., Reisler (1976), which, despite its title, is exclusively about male Mexican labor. 4 I use the term racial-ethnic to refer collectively to groups that have been socially constructed and constituted as racially as well as culturally distinct from European Americans and placed in separate legal statuses from "free whites" (c.f. Omi and Winant 1986). Historically, African Americans, Latinos, Asian Americans, and Native Americans were so constructed. Similarly, I have capitalized the word Black throughout this article to signify the racial-ethnic construction of that category.

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

ever, they either take for grantedor ignore women's unpaid household labor and fail to considerwhetherthis work might also be "raciallydivided." In short, the racial division of reproductive labor has been a missing Thispiece, I would contend,is key pieceof the picturein both literatures. to the distinct exploitation of women of color and is a source of both hierarchyand interdependence among white women and women of color. It is thus essentialto the developmentof an integratedmodel of race and gender,one that treats them as interlocking,ratherthan additive, systems. In this articleI presenta historicalanalysisof the simultaneousrace of reproductive and genderconstruction laborin the UnitedStates,based on comparative studyof women'swork in the South,the Southwest,and the FarWest.I arguethat reproductive labor has dividedalong racialas well as gender lines and that the specificcharacteristics of the division have variedregionallyand changedover time as capitalismhas reorganized reproductive labor, shiftingparts of it from the householdto the market. In the first half of the centuryracial-ethnicwomen were emlabor in white households, ployed as servantsto performreproductive women of onerous aspectsof that work; in relievingwhite middle-class the second half of the century,with the expansionof commodifiedservices (services turned into commercialproducts or activities), racialethnic women are disproportionately employed as service workers in institutional labor, settingsto carryout lower-level"public"reproductive while cleanerwhite collar supervisoryand lower professionalpositions are filled by white women. I will examinethe ways race and genderwere constructedaroundthe divisionof laborby sketchingchangesin the organization of reproductive labor since the early nineteenthcentury,presentinga case study of domestic serviceamong AfricanAmericanwomen in the South, Mexican Americanwomen in the Southwest,and JapaneseAmericanwomen in Californiaand Hawaii, and finally examiningthe shift to institutional servicework, focusingon raceand genderstratification in healthcareand the racial division of labor within the nursing labor force. Race and gender emerge as socially constructed,interlockingsystems that shape the materialconditions,identities,and consciousnesses of all women.
Historical changes in the organization of reproduction

The concept of reproductive labor originatedin KarlMarx's remark that every system of production involves both the production of the necessitiesof life and the reproductionof the tools and labor power

Autumn

1992

SIGNS

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

necessaryfor production(Marx and Engels 1969, 31). Recent elaborations of the concept grow out of Engels'sdictumthat the "determining of force in historyis, in the last resort,the productionand reproduction immediatelife." This has, he noted,"a two-fold character,on the one hand the productionof subsistenceand on the other the productionof human beings themselves"(Engels1972, 71). Although often equated with domesticlabor or definednarrowlyas referringto the renewalof has come to be more broadly labor power,the termsocial reproduction conceived,particularly by social historians,to referto the creationand recreationof people as culturaland social, as well as physical, beings (Ryan1981, 15). Thus, it involvesmental,emotional,and manuallabor and Laslett1986, 117). This labor can be organizedin myriad (Brenner ways-in and out of the household, as paid or unpaid work, creating exchange value or only use value-and these ways are not mutually of food, which can be done by exclusive.An exampleis the preparation a family memberas unwaged work in the household, by a servant as waged work in the household, or by a short-ordercook in a fast-food as waged work that generatesprofit for the employer.These restaurant forms exist contemporaneously. Priorto industrialization, however,both productionand reproduction were organizedalmost exclusivelyat the householdlevel. Womenwere responsiblefor most of what might be designatedas reproduction,but engagedin the productionof foodstuffs,cloththey were simultaneously ing, shoes, candles,soap, and other goods consumedby the household. With industrialization, productionof these basic goods graduallywas taken over by capitalist industry. Reproduction,however, remained of individualhouseholds.The ideologicalseplargelythe responsibility aration between men's "productive"labor and women's non-marketbased activitythat had evolvedat the end of the eighteenthcenturywas An idealizeddivisionof in the earlydecadesof the nineteenth. elaborated labor arose in which men's work was to follow productionoutside the home, while women's work was to remain centeredin the household (Boydston1990, esp. 46-48). Householdwork continuedto includethe productionof manygoods consumedby members(Smuts1959, 11-13; Kessler-Harris 1981), but as an expanding rangeof outside-manufactured focusedon householdwork becameincreasingly goods becameavailable, This idealizeddivision of labor was largely illusory for reproduction.5 andracial-ethnic families, households,including immigrant working-class in which men seldomearneda familywage; in these householdswomen
5 Capitalism, however, changed the nature of reproductive labor, which became more and more devoted to consumption activities, i.e., using wages to acquire necessities in the market and then processing these commodities to make them usable (see Weinbaum and Bridges 1976; and Luxton 1980).

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

and childrenwere forcedinto income-earning activitiesin and out of the home (Kessler-Harris 1982). In the second half of the twentieth century,with goods production almost completelyincorporatedinto the market, reproductionhas become the next major target for commodification.Aside from the tendency of capital to expand into new areas for profit making, the very conditions of life brought about by large-scalecommodityproduction have increasedthe need for commercialservices.As householdmembers spendmore of theirwakinghoursemployedoutsidethe home, they have less time and inclinationto provide for one another'ssocial and emotional needs. With the growth of a more geographicallymobile and urbanizedsociety,individualsand householdshave becomeincreasingly cut off from largerkinshipcircles,neighbors,and traditionalcommunities. Thus, as HarryBraverman notes, "The populationno longer relies upon social organizationin the form of family,friends,neighbors,community,elders,children,but with few exceptionsmust go to the market and only to the market,not only for food, clothing,and shelter,but also for recreation, for the careof the young,the old, the amusement, security, In timenot only the materialand serviceneedsbut sick, the handicapped. even the emotional patternsof life are channeledthroughthe market" (Braverman 1974, 276). Conditionsof capitalisturbanismalso have enlargedthe populationof those requiringdaily care and support:elderly and very young people, mentallyand physicallydisabledpeople, criminals, and other people incapableof fendingfor themselves.Becausethe care of such dependentsbecomesdifficultfor the "stripped-down" nuclear family or the atomized communityto bear, more of it becomes relegatedto institutionsoutside the family.6 The final phase in this process is what Braverman calls the "product cycle,"which "inventsnew productsand services,some of which become as the conditionsof modernlife changeand destroyalterindispensable natives"(Braverman 1974, 281). In manyareas(e.g., healthcare),we no longerhave choices outside the market.New servicesand productsalso alter the definitionof an acceptablestandardof living. Dependenceon the market is furtherreinforcedby what happenedearlierwith goods so that individualsno production,namely,an "atrophyof competence," longer know how to do what they formerlydid for themselves. As a resultof thesetendencies,an increasing rangeof serviceshas been removedwholly or partiallyfrom the householdand convertedinto paid servicesyieldingprofits.Today,activitiessuch as preparingand serving food (in restaurantsand fast-food establishments),caring for handi6 This is not to deny that familymembers, especiallywomen, still providethe bulk of care of dependents, but to point out that therehas been a markedincreasein institutionalizedcarein the secondhalf of the twentiethcentury.

Autumn 1992

SIGNS

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

capped and elderly people (in nursing homes), caring for children (in child-care centers), and providing emotional support, amusement, and companionship (in counseling offices, recreation centers, and health clubs) have become part of the cash nexus. In addition, whether impelled by a need to maintain social control or in response to pressure exerted by worker and community organizations, the state has stepped in to assume minimal responsibility for some reproductive tasks, such as child protection and welfare programs.7 Whether supplied by corporations or the state, these services are labor-intensive. Thus, a large army of low-wage workers, mostly women and disproportionately women of color, must be recruited to supply the labor. Still, despite vastly expanded commodification and institutionalization, much reproduction remains organized at the household level. Sometimes an activity is too labor-intensive to be very profitable. Sometimes households or individuals in them have resisted commodification. The limited commodification of child care, for example, involves both elements. The extent of commercialization in different areas of life is uneven, and the variation in its extent is the outcome of political and economic struggles (Brennerand Laslett 1986, 121; Laslett and Brenner 1989, 384). What is consistent across forms, whether commodified or not, is that reproductive labor is constructed as "female." The gendered organization of reproduction is widely recognized. Less obvious, but equally characteristic, is its racial construction: historically, racial-ethnic women have been assigned a distinct place in the organization of reproductive labor. Elsewhere I have talked about the reproductive labor racial-ethnic women have carried out for their own families; this labor was intensified as the women struggled to maintain family life and indigenous cultures in the face of cultural assaults, ghettoization, and a labor system that relegated men and women to low-wage, seasonal, and hazardous employment (Glenn 1985; 1986, 86-108; Dill 1988). Here I want to talk about two forms of waged reproductive work that racial-ethnic women have performed disproportionately: domestic service in private households and institutional service work. Domestic service as the racial division of reproductive labor Both the demand for household help and the number of women employed as servants expanded rapidly in the latter half of the nineteenth
of controlversusagency of varying viewson the relative 7 Fora discussion importance note thatprograms statewelfare in shaping policy,see Gordon(1990).PivenandCloward to defusepresandareintended havebeencreated only whenpoorpeoplehavemobilized and CloMovements surefor moreradical (Piven change(1971, 66). In theirPoorPeople's fromthe to win concessions the roleof working-class ward 1979),theydocument struggles see Abramovitz socialcontrolperspective, state.Fora feminist (1988).
6 SIGNS Autumn 1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

century(Chaplin1978). This expansionparalleledthe rise of industrial capital and the elaboration of middle-classwomen's reproductiveresponsibilities.Rising standardsof cleanliness,largerand more ornately of the home as a "havenin a furnishedhomes, the sentimentalization heartless world" (Lasch 1977), and the new emphasis on childhood and the mother's role in nurturing children all served to enlarge for reproductionat a time when middle-classwomen's responsibilities technology had done little to reduce the sheer physical drudgeryof
housework.8

womendid not challenge the gender-based By all accountsmiddle-class of theirreproductive divisionof laboror the enlargement responsibilities. Indeed, middle-classwomen-as readers and writers of literature;as membersand leadersof clubs, charitableorganizations, associations,reof the causeof and religiousrevivals;and as supporters formmovements, and Laslett abolition-helped to elaboratethe domestic code (Brenner 1986).9Feminists seekingan expandedpublicrole for womenarguedthat and moral qualitiesthat made women centersof the the same nurturant home shouldbe broughtto bearin publicservice.In the domesticsphere, the inequitable insteadof questioning genderdivisionof labor,theysought tasks onto moreoppressedgroupsof to sloughoff the more burdensome
women.10

PhyllisPalmerobservesthat at least throughthe firsthalf of the twentieth century, "most white middle class women could hire another a workingclasswoman, a womanof color, woman-a recentimmigrant, or all three-to performmuch of the hard labor of household tasks" (Palmer1987, 182-83). Domesticswere employedto clean house, launder and iron clothes,scrubfloors,and carefor infantsand children.They relieved their mistressesof the heavier and dirtier domestic chores.11 women were therebyfreedfor supervisory tasks and White middle-class for cultural,leisure, and volunteeractivity or, more rarely during this period, for a career.12
8 These developments are discussed in Degler (1980), Strasser (1982), Cowan (1983), and Dudden (1983, esp. 240-42). 9 See also Blair (1980); Epstein (1981); Ryan (1981); Dudden (1983); and Brenner and Laslett (1986). 10 See, e.g., Kaplan (1987). 11 Phyllis Palmer, in her Domesticity and Dirt, found evidence that mistresses and servants agreed on what were the least desirable tasks-washing clothes, washing dishes, and taking care of children on evenings and weekends-and that domestics were more likely to perform the least desirable tasks (1990, 70). 12 It may be worth mentioning the importance of unpaid cultural and charitable activities in perpetuating middle-class privilege and power. Middle-class reformers often aimed to mold the poor in ways that mirrored middle-class values but without actually altering their subordinate position. See, e.g., Sanchez (1990) for discussion of efforts of Anglo reformers to train Chicanas in domestic skills.

Autumn 1992

SIGNS

Glenn

RACIAL

DIVISION

OF REPRODUCTIVE

LABOR

Palmer thatthe use of domesticservants also helpedresolvecersuggests tain contradictions createdby the domesticcode. She notes that the early housewife confronted inconsistent of middletwentieth-century expectations classwomanhood: and"feminine virtue." domesticity Domesticity-defined as creating a warm, clean, and attractive home for husband and children-required hardphysicallabor and meantcontendingwith dirt. The virtuouswoman, however,was definedin terms of spirituality, rein the 1920s finement,and the denialof the physicalbody.Additionally, and 1930s thereemergeda new ideal of the modernwife as an intelligent and attractive companion.If the heavypartsof householdwork could be to paid help, the middle-class transferred housewifecould fulfillher domesticduties,yet distanceherselffromthe physicallaboranddirtandalso havetime for personaldevelopment(Palmer1990, 127-51). Who was to performthe "dirtywork"variedby region.In the Norththose who were Irishand east, European women,particularly immigrant German,constitutedthe majorityof domestic servantsfrom the midnineteenthcenturyto WorldWarI (Katzman1978, 65-70). In regions where there was a large concentrationof people of color, subordinaterace women formeda more or less permanentservantstratum.Despite in the compositionof the populationsand the mix of indusdifferences tries in the regions,there were importantsimilaritiesin the situationof Mexicansin the Southwest,AfricanAmericansin the South, and Japanese people in northernCaliforniaand Hawaii.Eachof thesegroupswas placedin a separatelegalcategoryfromwhites,excludedfromrightsand protectionsaccordedfull citizens. This severelylimited their ability to organize,competefor jobs, and acquirecapital(Glenn1985). The racial divisionof privatereproductive work mirrored this racialdualismin the legal, political, and economicsystems. In the South, AfricanAmericanwomen constitutedthe main and almost exclusiveservantcaste. Exceptin times of extremeeconomiccrisis, whites and Blacks did not compete for domestic jobs. Until the First WorldWar90 percentof all nonagriculturally employedBlackwomenin the South were employedas domestics.Even at the national level, servants and laundresses accountedfor close to half (48.4 percent)of nonIn the Southwest,especially in the states with the highest proprotions of Mexicans in the population-Texas, Colorado, and New Mexico-Chicanas were disproportionately concentratedin domestic service.14 In El Paso nearlyhalf of all Chicanasin the labor marketwere
13 U.S. Bureau of the Census 1933, chap. 3, "Color and Nativity of Gainful Workers," tables 2, 4, 6. For discussion of the concentration of African American women in domestic service, see Glenn (1985). 14 I use the terms Chicano, Chicana, and Mexican American to refer to both nativeborn and immigrant Mexican people/women in the United States.

agriculturally employed Black women in 1930.13

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

employedas servantsor laundressesin the early decadesof the century (Garcia1981, 76). In Denver,accordingto SarahDeutsch,perhapshalf of all householdshad at least one femalememberemployedas a domestic at some time, and if a woman becamea widow, she was almost certain to take in laundry (Deutsch 1987a, 147). Nationally, 39.1 percent of nonagriculturally employed Chicanas were servants or laundressesin
1930.15

In the FarWest-especially in Californiaand Hawaii,with theirlarge sex ratio madefemale populationsof Asianimmigrants-an unfavorable labor scarcein the late nineteenthand earlytwentiethcenturies.In contrast to the rest of the nation, the majorityof domesticservantsin California and Hawaii were men: in Californiauntil 1880 (Katzman1978, 55) and in Hawaii as late as 1920 (Lind 1951, table 1). The men were Asian--Chineseand laterJapanese.Chinesehouseboysand cooks were familiar figures in late nineteenth-century San Francisco;so too were Honolulu.After 1907 Japanesemale retainersin earlytwentieth-century Japanesewomen began to immigratein substantialnumbers,and they inheritedthe mantle of service in both Californiaand Hawaii. In the pre-WorldWarII years, close to half of all immigrantand native-born JapaneseAmericanwomen in the San FranciscoBay area and in Honolulu were employedas servantsor laundresses (U.S.Bureauof the Census 1932, table 8; Glenn 1986, 76-79). Nationally,excludingHawaii, 25.4 percent of nonagricultural Japanese American women workers were listed as servantsin 1930.16 In areaswhere racial dualismprevailed,being servedby membersof the subordinate in the dominant group was a perquisiteof membership to Elizabeth RaeTyson,an Anglowomanwho grewup group.According in El Paso in the early years of the century, "almost every AngloAmericanfamily had at least one, sometimestwo or three servants:a maid and laundress,and perhaps a nursemaidor yardman.The maid came in after breakfastand cleaned up the breakfastdishes, and very likelylast night'ssupperdishesas well; did the routinecleaning,washing and ironing,and afterthe familydinnerin the middleof the day,washed dishesagain, and then went home to performsimilarservicesin her own home" (Garcia1980, 327). In southwestcities, Mexican Americangirls were trainedat an early age to do domesticwork and girls as young as nine or ten were hired to clean house.17 In Hawaii, where the major social division was between the haole (Caucasian) planterclass and the largelyAsian plantationworkerclass,
15 U.S. Bureauof the Census1933. 16 Ibid.
17 Forpersonalaccountsof Chicanochildrenbeinginductedinto domesticservice, see Ruiz (1987a) and interviewof Josephine Turietta in Elsasser, and Tixier MacKenzie, y Vigil (1980, 28-35).

Autumn

1992

SIGNS

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

haole residents were required to employ one or more Chinese or Japanese servants to demonstrate their status and their social distance from those less privileged. Andrew Lind notes that "the literature on Hawaii, especially during the second half of the nineteenth century, is full of references to the open-handed hospitality of Island residents, dispensed by the everpresent maids and houseboys" (Lind 1951, 73). A public school teacher who arrived in Honolulu in 1925 was placed in a teacher's cottage with four other mainland teachers. She discovered a maid had already been hired by the principal: "A maid! None of us had ever had a maid. We were all used to doing our own work. Furthermore, we were all in debt and did not feel that we wanted to spend even four dollars a month on a maid. Our principal was quite insistent. Everyone on the plantation had a maid. It was, therefore, the thing to do" (Lind 1951, 76). In the South, virtually every middle-class housewife employed at least one African American woman to do cleaning and child care in her home. Southern household workers told one writer that in the old days, "if you worked for a family, your daughter was expected to, too" (Tucker 1988, 98). Daughters of Black domestics were sometimes inducted as children into service to baby-sit, wash diapers, and help clean (Clark-Lewis 1987, 200-201).18 White-skin privilege transcended class lines, and it was not uncommon for working-class whites to hire Black women for housework (Anderson and Bowman 1953). In the 1930s white women tobacco workers in Durham, North Carolina, could mitigate the effects of the "double day"-household labor on top of paid labor-by employing Black women to work in their homes for about one-third of their own wages (Janiewski 1983, 93). Black women tobacco workers were too poorly paid to have this option and had to rely on the help of overworked husbands, older children, Black women too old to be employed, neighbors, or kin. Where more than one group was available for service, a differentiated hierarchy of race, color, and culture emerged. White and racial-ethnic domestics were hired for different tasks. In her study of women workers in Atlanta, New Orleans, and San Antonio during the 1920s and 1930s, Julia Kirk Blackwelder reported that "anglo women in the employ of private households were nearly always reported as housekeepers, while Blacks and Chicanas were reported as laundresses, cooks or servants" (Blackwelder 1978, 349).19
18 See also life history accounts of Black domestics, such as that of Bolden (1976) and of Anna Mae Dickson by Wendy Watriss (Watriss 1984). 19 Blackwelder also found that domestics themselves were attuned to the racial-ethnic hierarchy among them. When advertising for jobs, women who did not identify themselves as Black overwhelmingly requested "housekeeping" or "governess" positions, whereas Blacks advertised for "cooking," "laundering," or just plain "domestic work."

10

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

In the Southwest, where Anglos considered Mexican or "Spanish" culture inferior, Anglos displayed considerable ambivalence about employing Mexicans for child care. Although a modern-day example, this statement by an El Paso businessman illustrates the contradictions in Anglo attitudes. The man told an interviewer that he and his wife were putting off parenthood because "the major dilemma would be what to do with the child. We don't really like the idea of leaving the baby at home with a maid ... for the simple reason if the maid is Mexican, the child may assume that the other person is its mother. Nothing wrong with Mexicans, they'd just assume that this other person is its mother. There have been all sorts of cases where the infants learned Spanish before they learned English. There've been incidents of the Mexican maid stealing the child and taking it over to Mexico and selling it" (Ruiz 1987b, 71). In border towns, the Mexican group was further stratified by Englishspeaking ability, place of nativity, and immigrant status, with nonEnglish-speaking women residing south of the border occupying the lowest rung. In Laredo and El Paso, Mexican American factory operatives often employed Mexican women who crossed the border daily or weekly to do domestic work for a fraction of a U.S. operative's wages (Hield 1984, 95; Ruiz 1987a, 64). The race and gender construction of domestic service Despite their preference for European immigrant domestics, employers could not easily retain their services. Most European immigrant women left service upon marriage, and their daughters moved into the expanding manufacturing, clerical, and sales occupations during the 1910s and twenties.20 With the flow of immigration slowed to a trickle during World War I, there were few new recruits from Europe. In the 1920s, domestic service became increasingly the specialty of minority-race women (Palmer 1990, 12). Women of color were advantageous employees in one respect: they could be compelled more easily to remain in service. There is considerable evidence that middle-class whites acted to ensure the domestic labor supply by tracking racial-ethnic women into domestic service and blocking their entry into other fields. Urban school systems in the Southwest tracked Chicana students into homemaking courses designed to prepare them for domestic service. The El Paso school board established a segregated school system in the 1880s that remained in place for the next thirty years; education for Mexican chilThis is not to say that daughters of European immigrants experienced great social mobility and soon attained affluence. The nondomestic jobs they took were usually low paying and the conditions of work often deplorable. Nonetheless, white native-born and immigrant women clearly preferred the relative freedom of industrial, office, or shop employment to the constraints of domestic service (see Katzman 1978, 71-72).
20

Autumn

1992

SIGNS

11

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

dren emphasized manual and domestic skills that would prepare them to work at an early age. In 1909 the Women's Civic Improvement League, an Anglo organization, advocated domestic training for older Mexican girls. Their rationale is explained by Mario Garcia: "According to the league the housegirls for the entire city came from the Mexican settlement and if they could be taught housekeeping, cooking and sewing, every American family would benefit. The Mexican girls would likewise profit since their services would improve and hence be in greater demand" (Garcia 1981, 113). The education of Chicanas in the Denver school system was similarly directed toward preparing students for domestic service and handicrafts. Sarah Deutsch found that Anglo women there persisted in viewing Chicanas and other "inferior-race" women as dependent, slovenly, and ignorant. Thus, they argued, training Mexican girls for domestic service not only would solve "one phase of women's work we seem to be incapable of handling" but it would simultaneously help raise the (Mexican) community by improving women's standard of living, elevating their morals, and facilitating Americanization (Deutsch 1987b, 736). One Anglo writer, in an article published in 1917 titled "Problems and Progress among Mexicans in Our Own Southwest," claimed, "When trained there is no better servant than the gentle, quiet Mexicana girl" (Romero 1988a, 16). In Hawaii, with its plantation economy, Japanese and Chinese women were coerced into service for their husbands' or fathers' employers. According to Lind, prior to World War II: It has been a usual practice for a department head or a member of the managerial staff of the plantation to indicate to members of his work group that his household is in need of domestic help and to expect them to provide a wife or daughter to fill the need. Under the conditions which have prevailed in the past, the worker has felt obligated to make a member of his own family available for such service, if required, since his own position and advancement depend upon keeping the goodwill of his boss. Not infrequently, girls have been prevented from pursuing a high school or college education because someone on the supervisory staff has needed a servant and it has seemed inadvisable for the family to disregard the claim. [Lind 1951, 77] Economic coercion also could take bureaucratic forms, especially for women in desperate straits. During the Depression, local officials of the federal Works Project Administration (WPA) and the National Youth Administration (NYA), programs set up by the Roosevelt administration to help the unemployed find work, tried to direct Chicanas and Blacks to
12 SIGNS Autumn 1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

domesticservicejobs exclusively(Blackwelder 1984, 120-22; Deutsch 1987a, 182-83). In Colorado, local officials of the WPA and NYA advocatedhousehold trainingprojectsfor Chicanas.GeorgeBickel, assistant state director of the WPA for Colorado, wrote: "The average Spanish-American girl on the NYA programlooks forwardto little save a life devoted to motherhoodoften under the most miserablecircumstances"(Deutsch1987a, 183). Givensuch an outlook, it made sense to providetrainingin domesticskills. Young Chicanasdisliked domestic service so much that slots in the programswent begging.Older women, especiallysingle mothersstruggling to support their families, could not afford to refuse what was offered.The cruel dilemmathat such women faced was poignantlyexpressedin one woman's letter to PresidentRoosevelt: My name is LulaGordon.I am a Negro woman. I am on the relief. I havethreechildren.I haveno husbandand no job. I haveworked hard ever since I was old enough. I am willing to do any kind of work becauseI haveto supportmyselfand my children.I was under the impressionthat the governmentor the W.P.A.would give the Physical[sic]fit reliefclientswork. I havebeenprayingfor that time to come. A lady,ElizabethRamsie,almostin my condition,told me she was going to try to get some work. I went with her.Wewent to the Court House here in San Antonio, we talked to a Mrs. Beckmon. Mrs. Beckmontold me to phone a Mrs. Coyle becauseshe wanted some one to clean house and cook for ($5) five dollars a week. Mrs. Beckmonsaid if I did not take the job in the Private home I would be cut off from everythingall together.I told her I was afraid to accept the job in the private home because I have for a government job andwhen it opens up I want to take registered it. She said that she was takingpeople off of the reliefand I haveto take the job in the privatehome or none.... I needwork and I will do anythingthe governmentgives me to do.... Will you please 1984, 68-69] give me some work. [Blackwelder women were similarlycompelledto acceptdomesJapaneseAmerican tic servicejobs when they left the internment campsin which they were World To leave War II. the campsthey had to have a imprisonedduring a and women were forced to take positions as and residence, job many live-in servantsin variousparts of the country.When women from the SanFrancisco thereafterthe campswere closed, agenBay areareturned cies set up to assist the returneesdirectedthem to domesticservicejobs. Becausethey had lost their homes and possessionsand had no savings, returneeshad to take whateverjobs were offered them. Some became
Autumn 1992 SIGNS 13

Glenn

RACIAL

DIVISION

OF REPRODUCTIVE

LABOR

live-in servants to secure housing, which was in short supply after the war. In many cases domestic employment became a lifelong career (Glenn 1986). In Hawaii the Japanese were not interned, but there nonetheless developed a "maid shortage" as war-related employment expanded. Accustomed to cheap and abundant household help, haole employers became increasingly agitated about being deprived of the services of their "mamasans." The suspicion that many able-bodied former maids were staying at home idle because their husbands or fathers had lucrative defense jobs was taken seriously enough to prompt an investigation by a university researcher.21 Housewives told their nisei maids it was the maids' patriotic duty to remain on the job. A student working as a live-in domestic during the war was dumbfounded by her mistress's response when she notified her she was leaving to take a room in the dormitory at the university. Her cultured and educated mistress, whom the student had heretofore admired, exclaimed with annoyance: "'I think especially in war time, the University should close down the dormitory.' Although she didn't say it in words, I sensed the implication that she believed all the (Japanese) girls should be placed in different homes, making it easier for the haole woman."22 The student noted with some bitterness that although her employer told her that working as a maid was the way for her to do "your bit for the war effort," she and other haole women did not, in turn, consider giving up the "conveniences and luxuries of pre-war Hawaii" as their bit for the war.23 The dominant group ideology in all these cases was that women of color-African American women, Chicanas, and Japanese American women-were particularly suited for service. These racial justifications ranged from the argument that Black and Mexican women were incapable of governing their own lives and thus were dependent on whites-making white employment of them an act of benevolence-to the argument that Asian servants were naturally quiet, subordinate, and accustomed to a lower standard of living. Whatever the specific content of the racial characterizations, it defined the proper place of these groups as in service: they belonged there, just as it was the dominant group's place to be served. David Katzman notes that "ethnic stereotyping was the stock in trade of all employers of servants, and it is difficult at times to figure out
21 Document Ma 24, Romanzo Adams Social Research Laboratory papers. I used these records when they were lodged in the sociology department;they are currentlybeing cataloged by the university archives and a finding aid is in process. 22 Ibid., document Ma 15, 5.

23

Ibid.

14

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

whether blacks and immigrants were held in contempt because they were servants or whether urban servants were denigrated because most of the servants were blacks and immigrants" (Katzman 1978, 221). Even though racial stereotypes undoubtedly preceded their entry into domestic work, it is also the case that domestics were forced to enact the role of the inferior. Judith Rollins and Mary Romero describe a variety of rituals that affirmed the subordination and dependence of the domestic; for example, employers addressed household workers by their first names and required them to enter by the back door, eat in the kitchen, and wear uniforms. Domestics understood they were not to initiate conversation but were to remain standing or visibly engaged in work whenever the employer was in the room. They also had to accept with gratitude "gifts" of discarded clothing and leftover food (Rollins 1985, chap. 5; Romero 1987). For their part, racial-ethnic women were acutely aware that they were trapped in domestic service by racism and not by lack of skills or intelligence. In their study of Black life in prewar Chicago, St. Clair Drake and Horace Cayton found that education did not provide African Americans with an entree into white collar work. They noted, "Colored girls are often bitter in their comments about a society which condemns them to the 'white folks' kitchen'" (Drake and Cayton 1962, 246). Thirty-five years later, Anna May Madison minced no words when she declared to anthropologist John Gwaltney: "Now, I don't do nothing for white women or men that they couldn't do for themselves. They don't do anything I couldn't learn to do every bit as well as they do it. But, you see, that goes right back to the life that you have to live. If that was the life I had been raised up in, I could be President or any other thing I got a chance to be" (Gwaltney 1980, 173). Chicana domestics interviewed by Mary Romero in Colorado seemed at one level to accept the dominant culture's evaluation of their capabilities. Several said their options were limited by lack of education and training. However, they also realized they were restricted just because they were Mexican. Sixty-eight-year-old Mrs. Portillo told Romero: "There was a lot of discrimination, and Spanish people got just regular housework or laundry work. There was so much discrimination that Spanish people couldn't get jobs outside of washing dishes-things like that" (Romero 1988b, 86). Similarly, many Japanese domestics reported that their choices were constrained because of language difficulties and lack of education, but they, too, recognized that color was decisive. Some nisei domestics had taken typing and business courses and some had college degrees, yet they had to settle for "school girl" jobs after completing their schooling. Mrs. Morita, who grew up in San Francisco and was graduated from high
Autumn 1992 SIGNS 15

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

school in the 1930s, bluntly summarizedher options: "In those days therewas no two ways about it. If you wereJapanese, you eitherworked curios'shop)wherethey sell those little junks,or in an art store ('oriental you workedas a domestic.... Therewas no Japanesegirl workingin an Americanfirm"(Glenn1986, 122). Hanna Nelson, anotherof Gwaltney'sinformants,took the analysis the coercionthat kept AfricanAmerican one step further;she recognized as one that allowed womenin domesticservice.Shesaw this arrangement white women to exploit Blackwomen economicallyand emotionallyand exposedBlackwomen to sexual assaultsby white men, often with white women'scomplicity.She says, "I am a woman sixty-oneyearsold and I was borninto this world with some talent.But I havedone the work that motherdid. It is not throughany failingof mine that my grandmother's this is so. The whites took my mother'smilk by force, and I havelived to hear a humancreatureof my sex try to force me by threatof hungerto give my milk to an able man. I have grown to womanhood in a world wherethe saneryou are, the madderyou are madeto appear"(Gwaltney 1980, 7). Race and genderconsciousness Hanna Nelson displays a consciousnessof the politics of race and gendernot found among white employers.Employers'and employees' ladifferentpositionswithin the divisionof reproductive fundamentally bor gave them differentinterests and perspectives.Phyllis Palmerdescribesthe problemsthe YWCAand other reformgroups encountered to establishvoluntarystandards andworkinghours when they attempted for live-in domesticsin the 1930s. White housewivesinvariablyargued against any "rigid"limitationof hours; they insistedon provisionsfor that would overrideany hour limits. Housewivessaw their emergencies as limitless,and apparentlyfelt there was no justiown responsibilities ficationfor boundarieson domestics'responsibilities. They did not acin theirpositions:they themselves difference knowledgethe fundamental gained status and privileges from their relationships with their husbands-relationships that dependedon the performanceof wifely duties.They expecteddomesticsto devote long hours and hardwork to privileges help them succeedas wives, without, however,commensurate and status.To challengethe inequitable genderdivisionof laborwas too so white housewivespushedthe dilemmaonto difficultand threatening, other women, holding them to the same high standardsby which they themselveswere imprisoned(Kaplan1987; Palmer1990). subSomedomesticworkerswere highlyconsciousof theirmistresses' to chaland condemned their to their husbands ordination unwillingness

16

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

lenge their husbands' authority. Mabel Johns, a sixty-four-year-old widow, told Gwaltney: I work for a woman who has a good husband;the devil is good to her,anyway.Now that woman could be a good personif she didn't think she could just do everythingand have everything.In this world whatsoeveryou get you will pay for. Now she is a grown woman, but she won't know that simplething.I don't thinkthere's anythingwrong with her mind, but she is greedy and she don't believein admittingthat she is greedy.Now you may say what you willormay[sic] about people beinggood to you, but therejust ain' a living soul in this world that thinks more of you than you do of yourself.... She'sa grown woman, but she have to keep accounts and her husbandtells her whetheror not he will let her do thusand-so or buy this or that. [Gwaltney1980, 167] Blackdomesticsare also consciousthat a white woman'sstatuscomes from her relationshipto a white man, that she gains privilegesfrom the that blindsher to her own oppression,and that she therefore relationship willingly participatesin and gains advantagesfrom the oppressionof racial-ethnic women. Nancy Whiteputs the matterpowerfullywhen she says, My motherusedto saythatthe blackwomanis thewhiteman'smule and the white woman is his dog. Now, she said that to say this: we do the heavywork and get beat whetherwe do it well or not. But the whitewomanis closerto the masterandhe patsthemon the head and lets them sleep in the house, but he ain' gon' treatneitherone likehe was dealingwith a person.Now, if I was to tell a whitewoman that,the firstthingshewould do is to callyou a niggerandthenshe'd be realnice to herhusbandso he would come out hereand beatyou for telling his wife the truth. [Gwaltney1980, 148] Ratherthan challengethe inequityin the relationship with their husbands, white women pushed the burden onto women with even less power. They could justify this only by denying the domestic worker's womanhood,by ignoringthe employee'sfamilyties and responsibilities. SusanTuckerfound that southernwhite women talked about their servants with affectionand expressedgratitudethat they sharedwork with the servantthat they would otherwisehaveto do alone. Yetthe sense of basedon genderthat the women expressedturnedout to be commonality one-way.Domesticworkersknew that employersdid not want to know

Autumn

1992

SIGNS

17

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

much about their home situations (Kaplan 1987, 96; Tucker 1988). Mostly, the employers did not want domestics' personal needs to interfere with serving them. One domestic wrote that her employer berated her when she asked for a few hours off to pay her bills and take care of pressing business (Palmer 1990, 74). Of relations between white mistresses and Black domestics in the period from 1870 to 1920, Katzman says that in extreme cases "even the shared roles of motherhood could be denied." A Black child nurse reported in 1912 that she worked fourteen to sixteen hours a day caring for her mistress's four children. Describing her existence as a "treadmill life," she reported that she was allowed to go home "only once in every two weeks, every other Sunday afternoon-even then I'm not permitted to stay all night. I see my own children only when they happen to see me on the streets when I am out with the children [of her mistress], or when my children come to the yard to see me, which isn't often, because my white folks don't like to see their servants' children hanging around their premises."24 While this case may be extreme, Tucker reports, on the basis of extensive interviews with southern African American domestics, that even among live-out workers in the 1960s, White women were also not noted for asking about childcare arrangements. All whites, said one black woman, "assume you have a mother, or an older daughter to keep your child, so it's all right to leave your kids." Stories of white employers not believing the children of domestics were sick, but hearing this as an excuse not to work, were also common. Stories, too, of white women who did not inquire of a domestic's family-even when that domestic went on extended trips with the family-were not uncommon. And work on Christmas morning and other holidays for black mothers was not considered by white employers as unfair. Indeed, work on these days was seen as particularly important to the job. [Tucker 1988, 99] The irony is, of course, that domestics saw their responsibilities as mothers as the central core of their identity. The Japanese American women I interviewed, the Chicana day workers Romero interviewed, and the African American domestics Bonnie Thornton Dill interviewed all emphasized the primacy of their role as mothers (Dill 1980; Glenn 1986; Romero 1988b). As a Japanese immigrant single parent expressed it, "My children come first. I'm working to upgrade my children." Another domestic, Mrs. Hiraoka, confided she hated household work but would
24 "More Slavery at the South: A Negro Nurse," from the Independent (1912), in Katzman and Tuttle (1982, 176-85, 179).

18

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

keep working until her daughtergraduatedfrom optometry school.25 Romero's day workers arrangedtheir work hours to fit around their children'sschool hours so that they could be there when needed. For domestics,then, workinghad meaningpreciselybecauseit enabledthem to providefor their children. Perhapsthe most universaltheme in domesticworkers'statementsis that they are working so their own daughterswill not have to go into domesticserviceand confrontthe same dilemmasof leavingtheir babies to work. A JapaneseAmericandomesticnoted, "I tell my daughtersall the time, 'As long as you get a steady job, stay in school. I want you to make get a good job, not like me.'That'swhat I alwaystell my daughters: sure you're not stuck."26 In a similarvein, PearlRunnertold Dill, "My main goal was I didn't want themto follow in my footstepsas far as working"(Dill 1980, 109). Domesticworkerswantedto protecttheirdaughters from both the hardships and the dangersthat working in white homes posed. A Black domestic told Drake and Cayton of her hopes for her daughters:"I hope they may be able to escape a life as a domesticworker,for I know too well the things that make a girl desperateon these jobs" (Drake and Cayton 1962, 246). Whenthey succeedin helpingtheirchildrendo betterthan they themselves did, domesticsmay considerthat the hardshipswere worthwhile. Lookingback, Mrs. Runneris able to say, "I reallyfeel that with all the that I went through,I feel happy and proud that I was able to struggling keephelpingmy children,that theylistenedand that they all went to high school. So when I look back,I reallyfeel proud,even thoughat timesthe work was very hard and I came home very tired. But now, I feel proud about it. They all got their education"(Dill 1980, 113). Domesticsthus have to grapple with yet another contradiction.They must confront, natureof the work they do to acknowledge,and convey the undesirable their children,as an object lesson and an admonition,and at the same time maintaintheir children'srespect and their own sense of personal worth and dignity(Dill 1980, 110). Whenthey successfully managethat beliefthat "you areyour contradiction, they refutetheirwhite employers' work" (Gwaltney1980, 174). The racial division of public reproductive labor As noted earlier,the increasingcommodification of social reproduction since WorldWarII has led to a dramaticgrowth in employmentby
25

1977.

From an interview conducted by the author in the San Francisco Bay area in

26 Ibid.

Autumn

1992

SIGNS

19

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

women in such areas as food preparation and service, health care services, child care, and recreational services. The division of labor in public settings mirrors the division of labor in the household. Racial-ethnic women are employed to do the heavy, dirty, "back-room" chores of cooking and serving food in restaurants and cafeterias, cleaning rooms in hotels and office buildings, and caring for the elderly and ill in hospitals and nursing homes, including cleaning rooms, making beds, changing bed pans, and preparing food. In these same settings white women are disproportionately employed as lower-level professionals (e.g., nurses and social workers), technicians, and administrative support workers to carry out the more skilled and supervisory tasks. The U.S. Census category of "service occupations except private household and protective services" roughly approximates what I mean by "institutional service work." It includes food preparation and service, health care service, cleaning and building services, and personal services.27 In the United States as a whole, Black and Spanish-origin women are overrepresented in this set of occupations; in 1980 they made up 13.7 percent of all workers in the field, nearly double their proportion (7.0 percent) in the work force. White women (some of whom were of Spanish origin) were also overrepresented, but not to the same extent, making up 50.1 percent of all "service" workers, compared with their 36 percent share in the overall work force. (Black and Spanish-origin men made up 9.6 percent, and white men, who were 50 percent of the work force, made up the remaining 27.5 percent.)28 Because white women constitute the majority, institutional service work may not at first glance appear to be racialized. However, if we look more closely at the composition of specific jobs within the larger category, we find clear patterns of racial specialization. White women are preferred in positions requiring physical and social contact with the public, that is, waiters/waitresses, transportation attendants, hairdressers/ cosmetologists, and dental assistants, while racial-ethnic women are preferred in dirty back-room jobs as maids, janitors/cleaners, kitchen workers, and nurse's aides.29
and the U.S. Bureauof the CensusdivideserviceocThe U.S. LaborDepartment and service," household," "protective "private cupationsinto threemajorcategories: In this discus"serviceoccupations exceptprivatehouseholdand protectiveservices." sion, "servicework"refersonly to the latter.I omit privatehouseholdworkers,who and protectiveserviceworkers,who includefirefighters havebeen discussed previously, well paid, and police:these jobs, in additionto beingmale dominatedand relatively carrysome degreeof authority, includingthe rightto use force. of the Census(1984), chap. D, "DetailedPopulation 8Computed from U.S. Bureau of table278: "DetailedOccupation Characteristics," pt. 1; "UnitedStatesSummary," Personsby Sex, Race and SpanishOrigin,1980.28." Employed 29Ibid.
27

20

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

As in the case of domesticservice,who does what varies regionally, following racial-ethniccaste lines in local economies. Racializationis clearest in local economies where a subordinaterace/ethnicgroup is sizable enough to fill a substantialportion of jobs. In southerncities, Blackwomen are twice as likelyto be employedin serviceoccupationsas white women. For example,in Atlantain 1980, 20.8 percentof African Americanwomen were so employed, comparedwith 10.4 percent of white women. Whilethey were less than one-quarter (23.9 percent)of all women workers, they were nearly two-fifths (38.3 percent)of women serviceworkers.In Memphis,25.9 percentof AfricanAmericanwomen comparedwith 10.2 percentof white women were in services;though they made up only a third (34.5 percent) of the female work force, African American women were nearly three-fifths(57.2 percent) of women employed in this field. In southwesterncities Spanish-origin women specialize in service work. In San Antonio, 21.9 percent of with 11.6 percentof womenwere so employed,compared Spanish-origin white women; in that city half (49.8 percent)of all non-Spanish-origin women serviceworkerswere Spanish-origin, while Anglos,who madeup two-thirds(64.0 percent)of the female work force, were a little over a third (36.4 percent) of those in the service category.In El Paso, 16.9 women were serviceworkers comparedwith percentof Spanish-origin 10.8 percentof Anglo women, and they made up two-thirds(66.1 percent) of those in service.Finally,in Honolulu,Asianand PacificIslanders constituted68.6 percentof the female work force, but 74.8 percentof those were in servicejobs. Overall,these jobs employed21.6 percentof all Asian and PacificIslanderwomen, comparedwith 13.7 percent of white non-Spanish-origin women.30
30 Figurescomputedfrom table279 in each of the state chaptersof the following: U.S. Bureauof the Census(1984), chap. D, "DetailedPopulationCharacteristics," pt. 6: "California"; pt. 15: "Illinois"; pt. 44: "Tennespt. 12: "Georgia"; pt. 13: "Hawaii"; The figuresfor Anglosin the Southwestare estimates,based see"; and pt. 45: "Texas." on the assumption that most "Spanish-origin" people are Mexican,and that Mexicans, are countedas whites. Specifically, the excess left after when given a racialdesignation, from the "sum"of white, Black,American the "total"is subtracted Indian/Eskimoand "Spanish-origin" is subtracted from the white fig/Aleut,Asianand PacificIslander, is countedas "Anglo." ure. The remainder Becauseof the way "Spanish-origin" crosscuts race (Spanish-origin individuals can be countedas white, Black,or any otherrace), I did not attemptto computefiguresfor Latinosor Anglosin citieswhereSpanish-origin in some unknownproportionbetweenBlack individuals are likelyto be more distributed and white. This would be the case, e.g., with the largePuertoRicanpopulationin New YorkCity.Thus I havenot attempted to computeLatinoversusAnglo data for New Yorkand Chicago.Note also that the meaningof white differsby locale and that the local termsAnglo and haole are not synonymous with white. The "white"categoryin HawaiiincludesPortuguese, who, becauseof theirhistoryas plantationlabor,are distinguishedfrom haoles in the local ethnicrankingsystem.The U.S. Censuscategorysystem of race/ethnicity. does not capturethe local construction

Autumn 1992

SIGNS

21

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Particularly striking is the case of cleaning and building services. This category-which includes maids, housemen, janitors, and cleaners-is prototypically "dirty work." In Memphis, one out of every twelve Black women (8.2 percent) was in cleaning and building services, and Blacks were 88.1 percent of the women in this occupation. In contrast, only one out of every 200 white women (0.5 percent) was so employed. In Atlanta, 6.6 percent of Black women were in this field-constituting 74.6 percent of the women in these jobs-compared with only 0.7 percent of white women. Similarly, in El Paso, 4.2 percent of Spanish-origin women (versus 0.6 percent of Anglo women) were in cleaning and building services-making up 90.0 percent of the women in this field. And in San Antonio the Spanish and Anglo percentages were 5.3 percent versus 1.1 percent, respectively, with Spanish-origin women 73.5 percent of women in these occupations. Finally, in Honolulu, 4.7 percent of Asian and Pacific Islander women were in these occupations, making up 86.6 percent of the total. Only 1.3 percent of white women were so employed.31

Frompersonalto structuralhierarchy
Does a shift from domestic service to low-level service occupations represent progress for racial-ethnic women? At first glance it appears not to bring much improvement. After domestic service, these are the lowest paid of all occupational groupings. In 1986 service workers were nearly two-thirds (62 percent) of workers in the United States earning at or below minimum wage.32 As in domestic service, the jobs are often parttime and seasonal, offer few or no medical and other benefits, have low rates of unionization, and subject workers to arbitrary supervision. The service worker also often performs in a public setting the same sorts of tasks that servants did in a private setting. Furthermore, established patterns of race/gender domination-subordination are often incorporated into the authority structure of organizations. Traditional gender-race etiquette shapes face-to-face interaction in the workplace. Duke University Hospital in North Carolina from its founding in 1929 adopted paternalistic policies toward its Black employees. Black workers were highly conscious of this, as evidenced by their references to "the plantation system" at Duke (Sacks 1988, 46).33 Still, service workers, especially those who have worked as domestics, are convinced that "public jobs" are preferable to domestic service. They appreciate not being personally subordinate to an individual employer
Computed from tables specified in ibid. The federal minimum wage was $3.35 in 1986. Over a quarter (26.0 percent) of all workers in these service occupations worked at or below this wage. See Mellor (1987, esp. 37). 33 Paternalism is not limited to southern hospitals; similar policies were in place at Montefiore Hospital in New York City. See Fink and Greenberg (1979).
31 32

22

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

and not havingto do "their"dirtywork on "their"property.Relations but they areembeddedin an and clientsarehierarchical, with supervisors impersonalstructuregovernedby more explicit contractualobligations and limits.Also importantis the presenceof a work groupfor sociability and support. Workplaceculture offers an alternativesystem of values workers from that imposed by managers(Benson1986).34Experienced to speed socializenewcomers,teachingthemhow to respondto pressures up work, to negotiatework loads, and to demandrespectfromsuperiors. While the isolated domestic finds it difficultto resist demeaningtreatment, the peer group in public settingsprovidesbackingfor individuals to stand up to the boss. That subordinationis usually not as direct and personal in public settings as in the privatehousehold does not mean, however,that race and gender hierarchyis diminishedin importance.Rather,it changes within organizational structures. Hierform, becominginstitutionalized archy is elaboratedthrough a detailed division of labor that separates conception from execution and allows those at the top to control the work process. Rankingis based ostensiblyon expertise,education,and formal credentials. The elaborationis especiallymarkedin technologically orientedorganizationsthat employlargenumbersof professionals,as is the case with health care institutions.Visual observationof any hospital reveals the hierarchical race and genderdivision of labor: at the top are the physicians, settingpolicy and initiatingwork for others; they are dispropormedicaltasks and tionatelywhite and male. Directlybelow, performing patient care as delegatedby physiciansand enforcinghospitalrules, are femaleand disprothe registered nurses(RNs), who are overwhelmingly nursesand often supervisedby portionatelywhite. Underthe registered them are the licensedpracticalnurses(LPNs),also femalebut disproportionatelywomen of color. At about the same level are the technologists and technicianswho carry out various tests and proceduresand the "administrative service"staff in the offices; these categoriestend to be female and white. Finally,at the bottom of the pyramidare the nurse's women of color; housekeepers and kitchenworkaides, predominantly women of color; and orderliesand cleaners,primaers, overwhelmingly rilymen of color.They constitutethe "hands"that performroutinework directedby others. The racialdivision of labor in nursing A study of stratification in the nursinglabor force illustratesthe race of publicreproductive and genderconstruction labor.At the top in terms
in Sacksand culturessupporting resistance 34 See also manyexamplesof workplace (1987). Remy (1984) and Lamphere
Autumn 1992 SIGNS 23

Glenn

RACIAL

DIVISION

OF REPRODUCTIVE

LABOR

of status, authority, and pay are the RNs, graduates of two-, three-, or four-year hospital or college-based programs. Unlike the lower ranks, registered nursing offers a career ladder. Starting as a staff nurse, a hospital RN can rise to head nurse, nursing supervisor, and finally, director of nursing. In 1980 whites were 86.7 percent of RNs even though they were only 76.7 percent of the population. The LPNs, who make up the second grade of nursing, generally have had twelve months' training in a technical institute or community college. The LPNs are supervised by RNs and may oversee the work of aides. Racial-ethnic workers constituted 23.4 percent of LPNs, with Blacks, who were 11.7 percent of the population, making up fully 17.9 percent. Below the LPNs in the hierarchy are the nurse's aides (NAs), who typically have on-the-job training of four to six weeks. Orderlies, attendants, home health aides, and patient care assistants also fall into this category. These workers perform housekeeping and routine caregiving tasks "delegated by an RN and performed under the direction of an RN or LPN." Among nurse's aides, 34.6 percent were minorities, with Blacks making up 27.0 percent of all aides.35 Nationally, Latinas were underrepresented in health care services but were found in nurse's aide positions in proportion to their numbers-making up 5.2 percent of the total. The lower two grades of nursing labor thus appear to be Black specialties. However, in some localities other women of color are concentrated in these jobs. In San Antonio, 48 percent of aides were Spanish-origin, while only 15.1 percent of the RNs were. Similarly, in El Paso, 61.5 percent of aides were Spanish-origin, compared with 22.8 percent of RNs. In Honolulu, Asian and Pacific Islanders who were 68.6 percent of the female labor force made up 72.3 percent of the NAs but only 45.7 percent of the
RNs.36

andtheraceandgender construction Familial of nursing. How symbolism

did the present ranking system and sorting by race/ethnic category in nursing come about? How did the activities of white nurses contribute to the structuring? And how did racial-ethnic women respond to constraints?
35 American in statusand authordifferences Nurses'Association1965, 6. Reflecting ity, RNs earn20-40 percentmorethan LPNsand 60-150 percentmorethan NAs of Labor1987a, 1987b). (U.S.Department 36 Forthe national level, see U.S. Bureauof the Census(1984), chap.D, "Detailed table278. Forstatisticson pt. 1: "UnitedStatesSummary," PopulationCharacteristics," RNs and aides in SanAntonio,El Paso,and Honolulu,see U.S. Bureauof the Census Characteristics," (1984), chap. D, "DetailedPopulation pt. 13: "Hawaii";and pt. 45: table279. "Texas,"

24

SIGNS

Autumn.

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

of nursinglabor can be tracedto the beginningsof The stratification organizednursingin the 1870s. However,until the 1930s gradingwas loose. A broad distinctionwas made between so-called trainednurses, who were graduatesof hospitalschools or collegiateprograms,and untrainednurses,referred to-often interchangeably-as "practical nurses," "hospital helpers,""nursingassistants,""nursingaides," or simply as "aides"(Canningsand Lazonik 1975; Reverby1987). Duringthis periodhealthwork in hospitalswas dividedbetweenmale physicians(patientdiagnosis and curing)and female nursingstaff (patient care) in a fashion analogousto the separatespheresprescribed for middle-classhouseholds. Nurses and physicianseach had primaryresponsibilityfor and authoritywithin theirown spheres,but nurseswere subject to the ultimate authority of physicians. The separation gave women power in a way that did not challengemale domination.Eva Gamarinikow likensthe position of the Britishnursingmatronto that of an upper-class woman in a Victorianhouseholdwho superviseda large household staff but was subordinateto her husband (Gamarinikow 1978). Taking the analogy a step further,Ann Game and Rosemary Pringle describe the pre-World War II hospital as operating under a system of controls based on familial symbolism. Physicianswere the authoritative fatherfigures,while trainednurseswere the mothersoverseeing the care of patients, who were viewed as dependentchildren. Studentnursesand practicalnurseswere, in this scheme,in the position of servants, expected to follow orders and subject to strict discipline (Gameand Pringle1983, 99-100). Like the middle-classwhite housewives who accepted the domestic ideology,white nursingleadersrarelychallengedthe familialsymbolism supportingthe genderdivision of labor in health care. The boldest advocated at most a dual-headedfamily (Reverby1987, 71-75). They acceded to the racial implicationsof the family metaphor as well. If nurseswere mothersin a family headed by white men, they had to be white. And, indeed, trained nursing was an almost exclusively white preserve.As SusanReverbynotes, "In 1910 and 1920, for example,less than 3% of the trainednursesin the United Stateswere black, whereas black women made up 17.6% and 24.0% respectivelyof the female workingpopulation"(Reverby1987, 71-75). The scarcityof Blackwomen is hardlysurprising. Nursingschools in the SouthexcludedBlacksaltogether, while northernschools maintained strict quotas. Typical was the policy of the New EnglandHospital for Women and Children,which by chartercould only admit "one Negro and one Jewishstudent"a year (Hine 1989, 6). Blackwomen who managed to become trained nurses did so through separateBlack training

Autumn

1992

SIGNS

25

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

schools and were usually restricted to serving Black patients, whether in "integrated" hospitals in the North or segregated Black hospitals in the South.37 White nursing leaders and administrators justified exclusion by appeals to racist ideology. Anne Bess Feeback, the superintendent of nurses for Henry Grady Hospital in Atlanta, declared that Negro women under her supervision had no morals: "They are such liars.... They shift responsibility whenever they can.... They quarrel constantly among themselves and will cut up each other's clothes for spite.... Unless they are constantly watched, they will steal anything in sight" (Hine 1985, 101). Perhaps the most consistent refrain was that Black women were deficient in the qualities needed to be good nurses: they lacked executive skills, intelligence, strength of character, and the ability to withstand pressure. Thus Margaret Butler, chief nurse in the Chicago City Health Department, contended that Black nurses' techniques were "inferior to that of the white nurses, they are not punctual, and are incapable of analyzing a social situation." Apparently Black nurses did not accept white notions of racial inferiority, for Butler also complains about their tendency "to organize against authority" and "to engage in political intrigue" (Hine 1989, 99). Another white nursing educator, Margaret Bruesche, suggested that although Black women lacked the ability to become trained nurses, they "could fill a great need in the South as a trained attendant, who would work for a lower wage than a fully trained woman" (Hine 1989, 101). Even those white nursing leaders sympathetic to Black aspirations agreed that Black nurses should not be put in supervisory positions because white nurses would never submit to their authority. Similar ideas about the proper place of "Orientals" in nursing were held by haole nursing leaders in pre-World War II Hawaii. White-run hospitals and clinics recruited haoles from the mainland, especially for senior nurse positions, rather than hiring or promoting locally trained Asian American nurses. This pattern was well known enough for a University of Hawaii researcher to ask a haole health administrator whether it was true that "oriental nurses do not reach the higher positions of the profession?" Mr. "C" confirmed this: "Well, there again it is a matter of qualification. There is a limit to the number of nurses we can produce
37 For accounts of Black women in nursing, see also Hine (1985) and Carnegie (1986). Hine (1989, chap. 7) makes it clear that Black nurses served Black patients not just because they were restricted but because they wanted to meet Black health care needs. Blacks were excluded from membership in two of the main national organizations for nurses, the National League of Nursing Education and the American Nurses' Association. And although they formed their own organizations such as the National Association of Colored Graduate Nurses and enjoyed the respect of the Black community, Black nurses remained subordinated within the white-dominated nursing profession.

26

SIGNS

Autumn

1992

RACIAL

DIVISION

OF REPRODUCTIVE

LABOR

Glenn

here.Forthatreasonwe haveto hirefromthe mainland.Localgirlscannot competewith the experienceof mainlandhaole girls. In orderto induce haole nursesherewe could not possiblyput themunderan orientalnurse becausethat would make them race consciousright at the start. And as I said before, Japanesedon't make good executives."38 Becauseof the racialcaste systemin Hawaii,JapaneseAmericanwomen who managed to get into nursingwere not seen as qualifiedor competentto do proof the Territorial fessionalwork. The chairman NursesAssociationnoted our local nurseswere looked down (upon) that "beforethe war (started), becausetheywere mostlyJapanese.... TheJapanesenursesfeel they can get alongbetterwith Mainlandnursesthanlocal haolenurses.Thatis true evenoutsideof the profession.I remember hearinga Hawaiianbornhaole dentistsay,'I was neverso shockedas when I saw a white manshineshoes when I first went to the Mainland.'Haoles here feel only orientalsand other non-haolesshould do menialwork."39 The systematicgradingof nursinglabor into threerankswas accomplishedin the 1930s and forties as physician-controlled hospital administrationsmoved to establish"soundbusiness"practicesto containcosts and consolidate physician control of health care.40High-tech medical and diagnosticprocedures providedan impetusfor ever-greater specialization. Hospitals adopted Tayloristprinciples of "scientificmanagement,"separating planningand technicaltasks from executionand manual labor. They began to hire thousands of subsidiaryworkers and createdthe licensedpracticalnurse, a position for a graduateof a oneyear technicalprogram, to perform routine housekeepingand patient care. With fewer discriminatory barriersand shorter trainingrequirements, LPN positionswere accessibleto women of color who wanted to becomenurses. The lowest level of nursingworkers,nurse'saides, also was definedin the 1930s, when the American Red Cross started offering ten-week coursesto trainaidesfor hospitals.This categoryexpandedrapidlyin the 1940s, doubling from 102,000 workers in 1940 to 212,000 in 1950 and Lazonik1975, 200-201). This occupationseemsto have (Cannings been designeddeliberately to make use of AfricanAmericanlabor in the
38 Document Nu21-I, p. 2, Romanzo Adams Research Laboratory papers, A1989006, box 17, folder 1. 39 Document NulO-I, p. 3, Romanzo Adams Research Laboratory papers, A1989006, box 17, folder 4. 40 This was one outcome of the protracted and eventually successful struggle waged by physicians to gain control over all health care. For an account of how physicians established hospitals as the main site for medical treatment and gained authority over "subsidiary" health occupations, see Starr (1982). For accounts of nurses' struggle for autonomy and their incorporation into hospitals, see Reverby (1987) and also Wagner (1980).

Autumn

1992

SIGNS

27

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

wake of labor shortages during and after World War II. A 1948 report on nursing told the story of how nurse's aides replaced the heretofore volunteer corps of ward attendants: "In response to this request for persons designated as nursing aides, the hospital discovered among the large Negro community a hitherto untapped reservoir of personnel, well above the ward attendant group in intelligence and personality" (Cannings and Lazonik 1975, 201). One reason for their superiority can be deduced: they often were overqualified. Barred from entry into better occupations, capable, welleducated Black women turned to nurse's aide work as an alternative to domestic service. In the meantime RNs continued their struggle to achieve professional status by claiming exclusive rights over "skilled" nursing work. Some nurses, especially rank-and-file general duty nurses, called for an outright ban on employing untrained nurses. Many leaders of nursing organizations, however, favored accepting subsidiary workers to perform housekeeping and other routine chores so that graduate nurses would be free for more professional work. Hospital administrators assured RNs that aides would be paid less and assigned non-nursing functions and that only trained nurses would be allowed supervisory roles. One administrator claimed that aide trainees were told repeatedly that "they are not and will not be nurses" (Reverby 1987, 194). In the end, the leaders of organized nursing accepted the formal stratification of nursing and turned their attention to circumscribing the education and duties of the lower grades to ensure their differentiation from "professional" nurses. Indeed, an RN arguing for the need to train and license practical nurses and laying out a model curriculum for LPNs warned: "Overtraining can be a serious danger. The practical nurse who has a course of over fifteen months (theory and practice) gets a false impression of her abilities and builds up the unwarranted belief that she can practice as a professional nurse" (Deming 1947, 26). Hospital administrators took advantage of race and class divisions and RNs' anxieties about their status to further their own agenda. Their strategy of co-opting part of the work force (RNs) and restricting the mobility and wages of another part (LPNs and NAs) undermined solidarity among groups that might otherwise have united around common interests. Nursing aides: Consciousness of race and gender. The hierarchy in health care has come to be justified less in terms of family symbolism and more in terms of bureaucratic efficiency. Within the new bureaucratic structures, race and gender ordering is inherent in the job definitions. The nurse's aide job is defined as unskilled and menial; hence, the women who do it are, too. Nurse's aides frequently confront a discrepancy,
28 SIGNS Autumn 1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

however,betweenhow theirjobs are defined(unskilledand subordinate) and what they actuallyare allowed or expectedto do (exerciseskill and judgment).Lillian Roberts'sexperiencesillustratethe disjunction.Assigned to the nursery,she was fortunateto work with a white southern RN who was willing to teach her. "I would ask her about all kinds of deformities that we would see in the nursery, the color of a baby,and why this was happeningand why the other was happening.And then I exploredwith her using my own analysisof things.SometimesI'd be right just in observingand putting some common sense into it. Beforelong, when the internswould come in to examinethe babies,I could tell them what was wrong with every baby. I'd have them lined up for them" (Reverby1979, 297-98). The expertise Roberts developed through observation,questioning, and deductionwas not recognized,however.Thirtyyears later Roberts still smartsfrom the injusticeof not being allowed to sit in on the shift reports:"They never dignifyyou with that. Even though it would help you give better care. There were limitationson what I could do" (Reverby 1979, 298-99). She had to assume a deferentialmanner when dealing with white medicalstudentsand personnel,even those who had much less experience than she had. Sometimesshe would be left in chargeof the nursery and "I'dget a whole messof new studentsin therewho didn'tknow what to do. I would very diplomaticallyhave to direct them, although they resentedto hell that I was both blackand a nurse'saide. But I had to do it in such a way that they didn't feel I was claimingto know more than was not they did" (Reverby1979, 298). One of her biggestfrustrations to allowed to advance. Roberts describes being get on-the-jobtraining the "box" she was in: "I couldn'thaveaffordedto go to nursingschool. I neededthe income,and you can'tjust quit a job and go to school. I was caughtin a box, and the salarywasn't big enoughto saveto go to school. And gettinginto the nursingschoolswas a realracistproblemas well. So therewas a combinationof many things.And I used to say, "'Whydoes this countryhaveto go elsewhereand get peoplewhen peoplelike myself want to do something?"' 1979, 299). Whenshe becamea union (Reverby her was to set up a programin New organizer, proudestaccomplishment York that allowed aides to be trainedon the job to become LPNs. While Roberts'sexperienceworking in a hospital was typical in the 1940s and 1950s, today the typicalaide is employedin a nursinghome, in a convalescenthome, or in home health care. In these settings,aides are the primarycaregivers.41 The demandfor theirservicescontinuesto
41 Forexample,it has been estimatedthat 80 percentof all patientcarein nursing homes is providedby nurse'saides (see Coleman1989, 5). In 1988, 1,559,000 persons

Autumn

1992

SIGNS

29

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

shifts out of hospitalsand into such setgrow as treatmentincreasingly tings. Thus, even though aides have lost ground to RNs in hospitals, which have reorganized nursingservicesto recreateRNs as generalists, aides are expected to remain among the fastest-growingoccupations throughthe end of the century(Sekcenski1981, 10-16).42 Whateverthe setting,aide work continuesto be a specialtyof racialethnic women. The work is seen as unskilledand subordinateand thus appropriateto their qualificationsand status. This point was brought home to TimothyDiamondduringthe trainingcoursehe attendedas the sole white male in a mostly Blackfemalegroupof trainees:"Welearned elementary biologyandhow we wereneverto do healthcarewithoutfirst andwe learnednot to ask questionsbut consultingsomeonein authority; to do as we were told. As one of the students, a black woman from Jamaicausedto joke, 'I can'tfigureout whetherthey'retryingto teachus to be nurses'aides or black women' " (Diamond1988, 40). What exactlyis the natureof the reproductive labor that these largely minorityand supposedlyunskilledaidesand assistants perform? Theydo most of the day-to-day,face-to-facework of caringfor the ill and disabled: helpingpatients dress or change gowns, taking vital signs (temperature,blood pressure,pulse), assistingpatients to shower or giving bed baths, emptying bedpans or assisting patients to toilet, changing sheets and keepingthe area tidy, and feedingpatientswho cannot feed themselves. Thereis much "dirty"work, such as cleaningup incontinent mental and emotional patients. Yet there is another,unacknowledged, dimensionto the work: listeningto the reminiscences of elderlypatients to help them hold on to their memory,comfortingfrightenedpatients about to undergosurgery,and providingthe only human contact some patientsget. This caringwork is largelyinvisible,and the skills required to do it are not recognizedas real skills.43 That these nurse'saides are performingreproductive labor on behalf of other women (and ultimatelyfor the benefitof households,industry, and the state) becomesclearwhen one considerswho would do it if paid workersdid not. Indeed,we confrontthat situationfrequently today, as hospitals reduce the length of patient stays to cut costs. Patients are released"quicker and sicker"(Sacks1988, 165). Thispolicy makessense
were employed as RNs, 423,00 as LPNs, 1,404,00 as nurse's aides, orderlies, and attendants, and 407,000 as health aides (U.S. Department of Labor 1989, table 22). Nurse's aides and home health care aides are expected to be the fastest-growing occupations through the 1990s, according to Silvestri and Lukasiewicz (1987, 59). 42 For a description of trends and projections to the year 2000, see Silvestri and Lukasiewicz (1987). 43 Feminists have pointed to the undervaluing of female-typed skills, especially those involved in "caring" work (see Rose 1986).

30

SIGNS

Autumn

1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

only if it is assumedthat patientshave someone to provideinterimcare, administer medication,preparemeals, and clean for them until they can care for themselves.If such a personexists, most likely it is a woman-a wife, mother,or sister.She may haveto take time off from her daughter, job or quit. Her unpaidlabortakesthe placeof the paid work of a nurse's aide or assistantand savesthe hospitallabor costs. Her labor is thereby appropriatedto ensure profit (Glazer 1988). Thus, the situation of women as unpaidreproductive workersat home is inextricably boundto that of women as paid reproductive workers. Conclusions and implications This article began with the observationthat the racial division of laborhas beenoverlookedin the separateliteratures on race reproductive and gender.The distinctexploitationof women of color and an important sourceof difference amongwomen havetherebybeenignored.How, a does historical though, analysisof the racial division of reproductive laborilluminatethe lives of women of color andwhite women?Whatare its implicationsfor concertedpolitical action? In order to tackle these questions,we need to addressa broaderquestion,namely,how does the of race and gender?Does it take us analysisadvanceour understanding the I additive models have criticized? beyond The social constructionof race and gender Tracing how race and gender have been fashioned in one area of women's work helps us understand them as socially constructed systems of relationships-including symbols, normative beliefs, and differences. This understanding is practices-organized aroundperceived an importantcounterto the universalizing in feministthought. tendencies When feministsperceivereproductive laboronly as gendered,they imply that domesticlaboris identicalfor all women and that it thereforecan be the basis of a common identityof womanhood. By not recognizingthe different relationshipswomen have had to such supposedly universal femaleexperiencesas motherhoodand domesticity, they risk essentializit as and natural. static, fixed, eternal, ing gender-treating They fail to take seriouslya basicpremiseof feministthought,that genderis a social construct. If race and gender are socially constructedsystems, then they must ariseat specificmomentsin particular circumstances and changeas these circumstanceschange. We can study their appearance,variation, and modification over time. I havesuggestedthat one vantagepoint for looking at their developmentin the United Statesis in the changingdivision of labor in local economies.A key site for the emergenceof conceptsof
Autumn 1992 SIGNS 31

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

genderedand racializedlabor has been in regionscharacterized by dual labor systems. As subordinate-race women withindual laborsystems,AfricanAmerican, Mexican American,and JapaneseAmericanwomen were drawn into domestic service by a combination of economic need, restricted opportunities,and educationaland employmenttrackingmechanisms. Once they were in service,their associationwith "degraded" labor affirmedtheir supposednaturalinferiority. Althoughideologiesof "race" and "racialdifference" justifyingthe dual labor system alreadywere in womanhoodwere inventedand place, specificideas about racial-ethnic enactedin everydayinteractionsbetweenmistressesand workers.Thus ideologiesof raceand genderwerecreatedandverifiedin dailylife (Fields 1982). Two fundamental elementsin the constructionof racial-ethnic womanhood were the notion of inherenttraits that suited the women for serviceand the denialof the women's identitiesas wives and mothersin theirown right.Employers accepteda cult of domesticitythat purported to elevate the status of women as mothers and homemakers,yet they madedemandson domesticsthat hampered themfromcarrying out these in theirown households.How could employersmaintain responsibilities such seeminglyinconsistentorientations? Racialideologywas criticalin it explainedwhy women of color were suited resolvingthe contradiction: for degradingwork. Racial characterizations effectivelyneutralizedthe racial-ethnicwoman's womanhood, allowing the mistress to be "unaware"of the domestic'srelationship to herown childrenand household. The exploitationof racial-ethnic women'sphysical,emotional,and mental work for the benefit of white households thus could be rendered invisiblein consciousnessif not in reality. labor from householdto market,faceWith the shift of reproductive to-face hierarchyhas been replacedby structuralhierarchy.In institutional settings, stratificationis built into organizationalstructures,including lines of authority, job descriptions, rules, and spatial and temporalsegregation.Distance between higher and lower ordersis ensured by structuralsegregation.Indeed, much routine service work is organizedto be out of sight: it takes place behind institutionalwalls where outsidersrarelypenetrate(e.g., nursinghomes, chroniccare facilkitchens),or at nightor othertimes ities), in backrooms (e.g., restaurant when occupantsare gone (e.g., in office buildingsand hotels). Workers may appreciatethis time and space segregationbecauseit allows them It also makes interactions. some autonomyand freedomfromdemeaning them and theirwork invisible,however.In this situation,moreprivileged women do not have to acknowledgethe workers or to confront the contradictionbetween sharedwomanhood and inequalityby race and
32 SIGNS Autumn 1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

class. Racialideology is not necessaryto explain or justifyexploitation, not for lack of racism, but becausethe justificationfor inequalitydoes not haveto be elaborated in specifically racialterms:insteadit can be cast in termsof differencesin training,skill, or education.4 Because they are socially constructed,race and gender systems are subject to contestationand struggle. Racial-ethnicwomen continually have challengedthe devaluationof their womanhood. Domestics often did so covertly.They learnedto dissemble,consciously"puttingon an act" while inwardlyrejectingtheiremployers'premisesand maintaining a separateidentity rooted in their families and communities.As noted institutional serviceworkerscan resistdemeaning treatment more earlier, openly because they have the support of peers. Minority-racewomen hospital workershave been in the forefrontof labor militancy,staging walkouts and strikesand organizingworkplaces.In both domesticservice and institutionalservicework, women have transcended the limitations of theirwork by focusingon longer-term goals, such as their children'sfuture. Beyondadditivemodels: Race and genderas interlocking systems As the foregoingexamplesshow, race and genderconstructsare inextricablyintertwined.Each develops in the context of the other; they cannotbe separated. This is important becausewhen we see reproductive labor only as gendered,we extract gender from its context, which includes other interactingsystemsof power. If we begin with genderseparatedout, then we have to put race and class back in when we consider women of color and working-classwomen. We thus end up with an additivemodel in which white women have only genderand women of color have genderplus race. The interlockis evident in the case studies of domesticworkers and nurse'saides.In the traditional middle-class of household,the availability female domestic labor buttressed white male cheap privilegeby perpetlaboras women'swork, sustaining the uatingthe conceptof reproductive illusion of a protectedprivatespherefor women and displacingconflict away from husband and wife to strugglesbetween housewife and domestic. The racialdivisionof labor also bolsteredthe genderdivisionof labor indirectlyby offeringwhite women a slightlymoreprivileged position in for on Rollins's noexchange acceptingdomesticity.Expanding Judith tion that white housewives gained an elevated self-identityby casting Black domesticsas inferiorcontrastfigures,PhyllisPalmersuggeststhe dependentposition of the middle-classhousewife made a contrasting
44 That is, the concentration of minorityworkersin lower-level jobs can be attributed to theirlack of "humancapital"-qualifications-needed for certainjobs.

Autumn

1992

SIGNS

33

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

figure necessary. A dualistic conception of women as "good" and "bad," long a part of western cultural tradition, provided ready-made categories for casting white and racial-ethnic women as oppositional figures (Davidoff 1979; Palmer 1990, 11, 137-39). The racial division of reproductive labor served to channel and recast these dualistic conceptions into racialized gender constructs. By providing them an acceptable self-image, racial constructs gave white housewives a stake in a system that ultimately oppressed them. The racial division of labor similarly protects white male privilege in institutional settings. White men, after all, still dominate in professional and higher management positions where they benefit from the paid and unpaid services of women. And as in domestic service, conflict between men and women is redirected into clashes among women. This displacement is evident in health care organizations. Because physicians and administrators control the work of other health workers, we would expect the main conflict to be between doctors and nurses over work load, allocation of tasks, wages, and working conditions. The racial division of nursing labor allows some of the tension to be redirected so that friction arises between registered nurses and aides over work assignments and supervision. In both household and institutional settings, white professional and managerial men are the group most insulated from dirty work and contact with those who do it. White women are frequently the mediators who have to negotiate between white male superiors and racial-ethnic subordinates. Thus race and gender dynamics are played out in a threeway relationship involving white men, white women, and women of color. Beyond difference: Race and gender as relational constructs Focusing on the racial division of reproductive labor also uncovers the relational nature of race and gender. By "relational" I mean that each is made up of categories (e.g., male/female, Anglo/Latino) that are positioned, and therefore gain meaning, in relation to each other (Barrett 1987). Power, status, and privilege are axes along which categories are positioned. Thus, to represent race and gender as relationally constructed is to assert that the experiences of white women and women of color are not just different but connected in systematic ways. The interdependence is easier to see in the domestic work setting because the two groups of women confront one another face-to-face. That the higher standard of living of one woman is made possible by, and also helps to perpetuate, the other's lower standard of living is clearly evident. In institutional service work the relationship between those who do the dirty work and those who benefit from it is mediated and buffered
34 SIGNS Autumn 1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

so the dependence of one group on the other by institutionalstructures, for its standardof living is not apparent.Nonetheless,interdependence exists, even if white women do not come into actualcontactwith women
of color.45

also recognizesthat white and racial-ethnic The notion of relationality women have differentstandpointsby virtueof their divergentpositions. This is an importantcorrectiveto feministtheoriesof genderedthought that posit universalfemale modes of thinkinggrowing out of common experiencessuch as domesticityand motherhood.When they portray reproductivelabor only as gendered, they assume there is only one standpoint-that of white women. Hence, the activitiesand experiences of middle-class women become generic"female"experiencesand activities, and those of other groups becomevariant,deviant,or specialized. In line with recentworks on AfricanAmerican,Asian American,and Latinafeministthought,we see that taking the standpointof women of on raceand gender color givesus a differentand morecriticalperspective Anzaldua Collins Domesticworkers (Garcia 1989; 1990; 1990.) systems in particular-because they directlyconfrontthe contradictions in their lives and those of theirmistresses-develop an acuteconsciousness of the nature of race and interlocking genderoppression. race and genderas Perhapsa less obvious point is that understanding relationalsystemsalso illuminatesthe lives of white Americanwomen. Whitewomanhoodhas been constructednot in isolation but in relation to that of women of color. Therefore,race is integralto white women's genderidentities.In addition,seeingvariationin racialdivisionof labor acrosstimein different regionsgivesus a morevariegated pictureof white womanhood.Whitewomen's lives have been lived in many middle-class in relationto varying has beenconstructed their"gender" circumstances; white womanhoodas others, not just to Blackwomen. Conceptualizing monolithicallydefinedin oppositionto men or to Blackwomen ignores complexityand variationin the experiencesof white women. Implicationsfor feministpolitics race and gender as relational, interlocking,socially Understanding constructedsystems affects how we strategizefor change. If race and genderare socially constructedratherthan being "real"referentsin the and challenged. materialworld, then they can be deconstructed Feminists now havemadeconsiderable stridesin deconstructing we need to gender; An focuson deconstructing and race initial simultaneously. step in gender that supportthe presentdivision this processis to expose the structures of labor and the constructionsof race and genderaroundit.
45

Elsa Barkley Brown pointed this out to me in a personal communication.

Autumn

1992

SIGNS

35

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Seeing race and gender as interlocking systems, however, alerts us to sources of inertia and resistance to change. The discussion of how the racial division of labor reinforced the gender division of labor makes clear that tackling gender hierarchy requires simultaneously addressing race hierarchy. As long as the gender division of labor remains intact, it will be in the short-term interest of white women to support or at least overlook the racial division of labor because it ensures that the very worst labor is performed by someone else. Yet, as long as white women support the racial division of labor, they will have less impetus to struggle to change the gender division of labor. This quandary is apparent in cities such as Los Angeles, which have witnessed a large influx of immigrant women fleeing violence and poverty in Latin America, Southeast Asia, and the Caribbean. These women form a large reserve army of low-wage labor for both domestic service and institutional service work. Anglo women who ordinarily would not be able to afford servants are employing illegal immigrants as maids at below-minimum wages (McConoway 1987). Not only does this practice diffuse pressure for a more equitable sharing of household work but it also recreates race and gender ideologies that justify the subordination of women of color. Having a Latino or Black maid picking up and cleaning after them teaches Anglo children that some people exist primarily to do work that Anglos do not want to do for themselves. Acknowledging the relational nature of race and gender and therefore the interdependence between groups means that we recognize conflicting interests among women. Two examples illustrate the divergence. With the move into the labor force of all races and classes of women, it is tempting to think that we can find unity around the common problems of "working women." With that in mind, feminist policymakers have called for expanding services to assist employed mothers in such areas as child care and elderly care. We need to ask, Who is going to do the work? Who will benefit from increased services? The historical record suggests that it will be women of color, many of them new immigrants, who will do the work and that it will be middle-class women who will receive the services. Not so coincidentally, public officials seeking to reduce welfare costs are promulgating regulations requiring women on public assistance to work. The needs of employed middle-class women and women on welfare might thus be thought to coincide: the needs of the former for services might be met by employing the latter to provide the services. The divergence in interest becomes apparent, however, when we consider that employment in service jobs at current wage levels guarantees that their occupants will remain poor. However, raising their wages so that they can actually support themselves and their children at a decent level would mean many middle-class women could not afford these services.
36 SIGNS Autumn 1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

A secondexampleof an issue that at firstblushappearsto bridgerace and ethnic lines is the continuingearningsdisparitybetween men and of women women. Becauseoccupationalsegregation,the concentration in low-paying,female-dominated occupations,stands as the major obstacle to wage equity, some feminist policymakershave embracedthe worth (Hartmann1985; Acker1989). This stratconceptof comparable egy calls for equalizingpay for "male" and "female" jobs requiring similarlevelsof skill and responsibility, even if differingin content.Comand differential parableworth acceptsthe validityof a job hierarchy pay based on "real"differencesin skills and responsibility. Thus, for exambetweennursesand pharmacists but leaves ple, it attacksthe differential intact the differentialbetweennursesand nurse'saides. Yet the division between "skilled"and "unskilled" jobs is exactly where the racial division typicallyfalls. To addressthe problemsof women of color service workerswould requirea fundamental attackon the conceptof a hierarbetween chy of worth; it would call for flatteningthe wage differentials highest-and lowest-paidranks.A claim would have to be made for the of skill or responsibility. rightof all workersto a living wage, regardless These examplessuggestthat forginga politicalagendathat addresses the universalneeds of women is highly problematicnot just because women'sprioritiesdifferbut becausegains for some groupsmay require a corresponding loss of advantageandprivilegefor others.As the history of the racial division of reproductive labor reveals,conflict and contestation among women over definitionsof womanhood, over work, and over the conditions of family life are part of our legacy as well as the currentreality.This does not mean we give up the goal of concerted struggle.It meanswe give up tryingfalselyto harmonizewomen's interests. Appreciating the ways raceand genderdivisionof laborcreatesboth and interdependence may be a betterway to reach an underhierarchy of of women's lives. the interconnectedness standing Departmentsof Ethnic Studiesand Women'sStudies Universityof California,Berkeley References
Mimi. 1988. Regulating the Livesof Women: SocialWelfare Abramovitz, Policy from ColonialTimesto the Present.Boston:South End Press. Acker,Joan. 1989. Doing ComparableWorth:Gender,Class, and Pay Equity. Press. Philadelphia: TempleUniversity

Romanzo. Research of Hawaii Social ArAdams, Laboratory papers. University Manoa. chives,
Autumn 1992 SIGNS

37

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

American Nurses' Association. 1965. Health Occupations Supportive to Nursing. New York: American Nurses' Association. Anderson, C. Arnold, and Mary Jean Bowman. 1953. "The Vanishing Servant and the Contemporary Status System of the American South." American Journal of Sociology 59:215-30. Anzalduia,Gloria. 1990. Making Face, Making Soul-Haciendo Caras: Creative Critical Perspectives by Women of Color. San Francisco: Aunt Lute Foundation. Barrera, Mario. 1979. Race and Class in the Southwest: A Theory of Racial Inequality. Notre Dame, Ind., and London: University of Notre Dame Press. Barrett, Michele. 1980. Women's Oppression Today: Problems in Marxist Feminist Analysis. London: Verso. .1987. "The Concept of 'Difference.' " Feminist Review 26(July):29-41. Benson, Susan Porter. 1986. Counter Cultures: Saleswomen, Customers, and Managers in American Department Stores, 1890-1940. Urbana and Chicago: University of Illinois Press. Benston, Margaret. 1969. "The Political Economy of Women's Liberation." Monthly Review 21(September):13-27. Blackwelder, Julia Kirk. 1978. "Women in the Work Force: Atlanta, New Orleans, and San Antonio, 1930 to 1940." Journal of Urban History 4(3):33158, 349. . 1984. Women of the Depression: Caste and Culture in San Antonio, 1929-1939. College Station: Texas A&M University Press. Blair, Karen. 1980. The Clubwoman as Feminist: True Womanhood Redefined, 1868-1914. New York: Holmes & Meier. Blauner, Robert. 1972. Racial Oppression in America. Berkeley: University of California Press. Bolden, Dorothy. 1976. "Forty-two Years a Maid: Starting at Nine in Atlanta." In Nobody Speaks for Me! Self-Portraits of American Working Class Women, ed. Nancy Seifer. New York: Simon & Schuster. Boydston, Jeanne. 1990. Home and Work: Housework, Wages, and the Ideology of Labor in the Early Republic. New York: Oxford University Press. Braverman,Harry. 1974. Labor and Monopoly Capital: The Degradation of Labor in the Twentieth Century.New York and London: Monthly Review Press. Brenner, Johanna, and Barbara Laslett. 1986. "Social Reproduction and the Family."In Sociology, from Crisis to Science? Vol. 2, The Social Reproduction of Organization and Culture, ed. Ulf Himmelstrand, 116-31. London: Sage. Brown, Elsa Barkley. 1989. "Womanist Consciousness: Maggie Lena Walker and the Independent Order of Saint Luke." Signs: Journal of Women in Culture and Society 14(3):610-33. Cannings, Kathleen, and William Lazonik. 1975. "The Development of the Nursing Labor Force in the United States: A Basic Analysis." International Journal of Health Sciences 5(2):185-216. Carnegie, Mary Elizabeth. 1986. The Path We Tread: Blacks in Nursing, 18541954. Philadelphia: Lippincott. Chaplin, David. 1978. "Domestic Service and Industrialization." Comparative Studies in Sociology 1:97-127.

38

SIGNS

Autumn 1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

Clark-Lewis, Elizabeth. 1987. "This Work Had an End: African American Domestic Workers in Washington, D.C., 1910-1940." In "To Toil the Livelong Day": America's Women at Work, 1780-1980, ed. Carole Groneman and Mary Beth Norton. Ithaca, N.Y.: Cornell University Press. Coleman, Barbara. 1989. "States Grapple with New Law."AARP News Bulletin, 30(2):4-5. Collins, Patricia Hill. 1986. "Learning from the Outsider Within: The Sociological Significance of Black Feminist Thought." Social Problems 33(6):14-32. . 1990. Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment. New York: Allen & Unwin. Cowan, Ruth Schwartz. 1983. More Work for Mother: The Ironies of Household Technology from the Open Hearth to the Microwave. New York: Basic. Davidoff, Lenore. 1979. "Class and Gender in Victorian England: The Diaries of Arthur J. Munby and Hannah Cullwick." Feminist Studies 5(Spring): 86-114. Degler, Carl N. 1980. At Odds: Women and the Family in America from the Revolution to the Present. New York: Oxford University Press. Deming, Dorothy. 1947. The Practical Nurse. New York: Commonwealth Fund. Deutsch, Sarah. 1987a. No Separate Refuge: Culture, Class, and Gender on an Anglo-Hispanic Frontier in the American Southwest, 1880-1940. New York: Oxford University Press. .1987b. "Women and Intercultural Relations: The Case of Hispanic New Mexico and Colorado." Signs 12(4):719-39. Diamond, Timothy. 1988. "Social Policy and Everyday Life in Nursing Homes: A Critical Ethnography." In The Worth of Women's Work: A Qualitative Synthesis, ed. Anne Statham, Eleanor M. Miller, and Hans O. Mauksch. Albany, N.Y.: SUNY Press. Dill, Bonnie Thornton. 1980. "The Means to Put My Children Through: Childrearing Goals and Strategies among Black Female Domestic Servants." In The Black Woman, ed. La Frances Rodgers-Rose. Beverly Hills and London: Sage. .1988. "Our Mothers' Grief: Racial Ethnic Women and the Maintenance of Families."Journal of Family History 12(4):415-31. Drake, St. Clair, and Horace Cayton. (1945) 1962. Black Metropolis: A Study of Negro Life in a Northern City, vol. 1. New York: Harper Torchbook. Dudden, Faye E. 1983. Serving Women: Household Service in Nineteenth Century America. Middletown, Conn.: Wesleyan University Press. Elsasser, Nan, Kyle MacKenzie, and Yvonne Tixier y Vigil. 1980. Las Mujeres: Conversations from a Hispanic Community. Old Westbury,N.Y: FeministPress. Engels, Friedrich. 1972. The Origins of the Family, Private Property and the State. New York: International Publishers. Epstein, Barbara. 1981. The Politics of Domesticity: Women, Evangelism, and Temperance in Nineteenth Century America. Middletown, Conn.: Wesleyan University Press. Fields, Barbara. 1982. "Ideology and Race in American History." In Region, Race, and Reconstruction: Essays in Honor of C. Vann Woodward, ed. J. Morgan Kousser and James M. McPherson. New York: Oxford University Press.

Autumn 1992

SIGNS

39

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Fink, Leon, and Brian Greenberg. 1979. "Organizing Montefiore: Labor Militancy Meets a Progressive Health Care Empire." In Health Care in America: Essays in Social History, ed. Susan Reverby and David Rosner. Philadelphia: Temple University Press. Fox, Bonnie, ed. 1980. Hidden in the Household: Women's Domestic Labour under Capitalism. Toronto: Women's Press. Gamarinikow, Eva. 1978. "Sexual Division of Labour: The Case of Nursing." In Feminism and Materialism: Women and Modes of Production, ed. Annette Kuhn and Ann-Marie Wolpe, 96-123. London: Routledge & Kegan Paul. Game, Ann, and Rosemary Pringle. 1983. Gender at Work. Sydney: Allen & Unwin. Garcia, Alma. 1989. "The Development of Chicana Feminist Discourse, 19701980." Gender and Society 3(2):217-38. Garcia, Mario T. 1980. "The Chicana in American History: The Mexican Women of El Paso, 1880-1920: A Case Study." Pacific Historical Review 49(2):315-39. .1981. Desert Immigrants: The Mexicans of El Paso, 1880-1920. New Haven, Conn.: Yale University Press. Glazer, Nona. 1988. "Overlooked, Overworked: Women's Unpaid and Paid Work in the Health Services' 'Cost Crisis,"' International Journal of Health Services 18(2):119-37. Glenn, Evelyn Nakano. 1985. "Racial Ethnic Women's Labor: The Intersection of Race, Gender and Class Oppression." Review of Radical Political Economy 17(3):86-108. . 1986. Issei, Nisei, Warbride:Three Generations of Japanese American Women in Domestic Service. Philadelphia: Temple University Press. Gordon, Linda. 1990. "The New Feminist Scholarship on the Welfare State." In Women, the State, and Welfare,ed. Linda Gordon, 9-35. Madison: University of Wisconsin Press. Gwaltney, John, ed. 1980. Drylongso: A Self-Portrait of Black America. New York: Random House. Hartmann, Heidi I., ed. 1985. Comparable Worth: New Directions for Research. Washington, D.C.: National Academy Press. Hield, Melissa. 1984. "Women in the Texas ILGWU, 1933-50." In Speaking for Ourselves: Women of the South, ed. Maxine Alexander, 87-97. New York, Pantheon. Hine, Darlene Clark, ed. 1985. Black Women in the Nursing Profession: A Documentary History. New York: Pathfinder. . 1989. Black Women in White: Racial Conflict and Cooperation in the Nursing Profession, 1890-1950. Bloomington: Indiana University Press. Janiewski, Delores. 1983. "Flawed Victories: The Experiences of Black and White Women Workers in Durham during the 1930s." In Decades of Discontent: The Women's Movement, 1920-1940, ed. Lois Scharf and Joan M. Jensen, 85-112. Westport, Conn., and London: Greenwood. Kaplan, Elaine Bell. 1987. "'I Don't Do No Windows': Competition between the Domestic Worker and the Housewife." In Competition: A Feminist Taboo?" ed. Valerie Miner and Helen E. Longino. New York: Feminist Press at CUNY.
40 SIGNS Autumn 1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

Katzman, David M. 1978. Seven Days a Week: Women and Domestic Service in Industrializing America. New York: Oxford University Press. Katzman, David M., and William M. Tuttle, Jr., eds. 1982. Plain Folk: The Life Stories of Undistinguished Americans. Urbana and Chicago: University of Illinois Press. Kessler-Harris, Alice. 1981. Women Have Always Worked: A Historical Overview. Old Westbury, N.Y.: Feminist Press. . 1982. Out to Work: A History of Wage-earning Women in the United States. New York: Oxford University Press. King, Deborah K. 1988. "Multiple Jeopardy, Multiple Consciousness: The Context of a Black Feminist Ideology." Signs 14(1):42-72. Lamphere, Louise. 1987. From Working Daughters to Working Mothers: Immigrant Women in a New England Industrial Community. Ithaca, N.Y.: Cornell University Press. Lasch, Christopher. 1977. Haven in a Heartless World: The Family Besieged. New York: Basic. Laslett, Barbara, and Johanna Brenner. 1989. "Gender and Social Reproduction: Historical Perspectives."Annual Review of Sociology 15:381-404. Lind, Andrew. 1951. "The Changing Position of Domestic Service in Hawaii." Social Process in Hawaii 15:71-87. Luxton, Meg. 1980. More than a Labour of Love: Three Generations of Women's Work in the Home. Toronto: Women's Press. McConoway, Mary Jo. 1987. "The Intimate Experiment." Los Angeles Times Magazine, February 19, 18-23, 37-38. Marx, Karl, and Friedrich Engels. 1969. Selected Works, vol. 1. Moscow: Progress. Mellor, Earl F 1987. "Workers at the Minimum Wage or Less: Who They Are and the Jobs They Hold." Monthly Labor Review, July, 34-38. Omi, Michael, and Howard Winant. 1986. Racial Formation in the United States. New York: Routledge. Palmer, Phyllis. 1987. "Housewife and Household Worker: Employer-Employee Relations in the Home, 1928-1941." In "To Toil the Livelong Day": America's Women at Work, 1780-1980, ed. Carole Groneman and Mary Beth Norton, 179-95. Ithaca, N.Y.: Cornell University Press. . 1990. Domesticity and Dirt: Housewives and Domestic Servants in the United States, 1920-1945. Philadelphia: Temple University Press. Piven, Frances Fox, and Richard A. Cloward. 1971. Regulating the Poor: The Functions of Public Welfare. New York: Pantheon. . 1979. Poor People's Movements: Why They Succeed, How They Fail. New York: Pantheon. Reisler, Mark. 1976. By the Sweat of Their Brow: Mexican Immigrant Labor in the United States, 1900-1940. Westport, Conn.: Greenwood. Reverby, Susan M. 1979. "From Aide to Organizer: The Oral History of Lillian Roberts." In Women of America: A History, ed. Carol Ruth Berkin and Mary Beth Norton. Boston: Houghton Mifflin. . 1987. Ordered to Care: The Dilemma of American Nursing, 18501945. Cambridge: Cambridge University Press. Autumn 1992 SIGNS 41

Glenn

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Rollins, Judith. 1985. Between Women: Domestics and Their Employers. Philadelphia: Temple University Press. Romero, Mary. 1987. "Chicanas Modernize Domestic Service." Unpublished manuscript. . 1988a. "Day Work in the Suburbs: The Work Experience of Chicana Private Housekeepers." In The Worth of Women's Work: A Qualitative Synthesis, ed. Anne Statham, Eleanor M. Miller, and Hans O. Mauksch, 77-92. Albany: SUNY Press. . 1988b. "Renegotiating Race, Class and Gender Hierarchies in the Everyday Interactions between Chicana Private Household Workers and Employers." Paper presented at the 1988 meetings of the Society for the Study of Social Problems, Atlanta. Rose, Hilary. 1986. "Women's Work: Women's Knowledge." In What Is Feminism? ed. Juliet Mitchell and Ann Oakley, 161-83. Oxford: Basil Blackwell. Ruiz, Vicki L. 1987a. "By the Day or the Week: Mexicana Domestic Workers in El Paso." In Women on the U.S.-Mexico Border: Responses to Change, ed. Vicki L. Ruiz and Susan Tiano, 61-76. Boston: Allen & Unwin. . 1987b. "Oral History and La Mujer: The Rosa Guerrero Story." In Women on the U.S.-Mexico Border: Responses to Change, ed. Vicki L. Ruiz and Susan Tiano, 219-32. Boston: Allen & Unwin. Ryan, Mary P. 1981. Cradle of the Middle Class: The Family in Oneida County, New York, 1790-1865. Cambridge: Cambridge University Press. Sacks, Karen Brodkin. 1988. Caring by the Hour: Women, Work, and Organizing at Duke Medical Center. Urbana and Chicago: University of Illinois Press. Sacks, Karen Brodkin, and Dorothy Remy, eds. 1984. My Troubles Are Going to Have Trouble with Me: Everyday Trials and Triumphs of Women Workers. New Brunswick, N.J.: Rutgers University Press. Sanchez, George J. 1990. "'Go after the Women': Americanization and the Mexican Immigrant Woman, 1915-1929." In Unequal Sisters: A Multicultural Reader in Women's History, ed. Ellen Carol DuBois and Vicki L. Ruiz, 25063. New York: Routledge. Secombe, Wally. 1974. "The Housewife and Her Labour under Capitalism." New Left Review 83(January-February):3-24. Sekcenski, Edward S. 1981. "The Health Services Industry: A Decade of Expansion." Monthly Labor Review (May):10-16. Silvestri, George T., and John M. Lukasiewicz. 1987. "A Look at Occupational Employment Trends to the Year 2000." Monthly Labor Review (September): 46-63. Smuts, Robert W. 1959. Women and Work in America. New York: Schocken. Sokoloff, Natalie J. 1980. Between Money and Love: The Dialectics of Women's Home and Market Work. New York: Praeger. Starr, Paul. 1982. The Social Transformation of American Medicine. New York: Basic. Strasser, Susan. 1982. Never Done: A History of American Housework. New York: Pantheon. Tucker, Susan. 1988. "The Black Domestic in the South: Her Legacy as Mother and Mother Surrogate." In Southern Women, ed. Carolyn Matheny Dillman, 93-102. New York: Hemisphere. 42 SIGNS Autumn 1992

RACIAL DIVISION

OF REPRODUCTIVE

LABOR

Glenn

U.S. Bureau of the Census. 1932. Fifteenth Census of the United States: 1930, Outlying Territoriesand Possessions. Washington, D.C.: Government Printing Office. . 1933. Fifteenth Census of the United States: 1930, Population. Vol.5, General Report on Occupations. Washington,D.C.: GovernmentPrintingOffice. . 1984. Census of the Population, 1980. Vol. 1, Characteristics of the Population. Washington, D.C.: Government Printing Office. U.S. Department of Labor. 1987a. Industry Wage Survey: Hospitals, August 1985. Bureau of Labor Statistics Bulletin 2273. Washington, D.C.: Government Printing Office. . 1987b. Industry Wage Survey: Nursing and Personal Care Facilities, September 1985. Bureau of Labor Statistics Bulletin 2275. Washington, D.C.: Government Printing Office. . 1989. Employment and Earnings, January 1989. Bureau of Labor Statistics Bulletin. Washington, D.C.: Government Printing Office. Wagner, David. 1980. "The Proletarianization of Nursing in the United States, 1932-1945." International Journal of Health Services 10(2):271-89. Watriss, Wendy. 1984. "It's Something Inside You." In Speaking for Ourselves: Women of the South, ed. Maxine Alexander. New York: Pantheon. Weinbaum, Batya, and Amy Bridges. 1976. "The Other Side of the Paycheck." Monthly Review 28:88-103.

Autumn 1992

SIGNS

43